Virology and Immunology

P345 Immune recovery in acute and chronic HIV infection and the impact of thymic stromal lymphopoietin Gelpi, M*; Hartling, H; Thorsteinsson, K; Gerstoft, J; Ullum, H; Nielsen, S (Copenhagen, Denmark) P347 Car diovascular risk in HIV-positive subjects: analyses of T-cell phenotype and CD49d expression Zingaropoli, M*; D’Abramo, A; Iannetta, M; Oliva, A; d’Ettorre, G; Lichtner, M; Mastroianni, C; Ciardi, M; Vullo, V (Rome, Italy) P348 Baseline myeloid and lymphoid activation markers can predict time to viral load reduction under 50 copies/mL and CD4 recovery, respectively, after highly-active antiretroviral therapy initiation Iannetta, M*; Lichtner, M; Rossi, R; Savinelli, S; Vita, S; Mascia, C; Zuccalà, P; Marocco, R; Zingaropoli, M; Ciardi, M; d’Ettorre, G; Mastroianni, C; Vullo, V (Rome, Italy) P349 Impact of oestrogen plasma levels in modulation of immune activation among HIV-infected women and men undergoing successful antiretroviral therapy Marocco, R*; Lichtner, M; Tieghi, T; Belvisi, V; Pozzetto, I; Mascia, C; Zuccalà, P; Rossi, R; Mengoni, F; Mastroianni, C (Latina, Italy) P351 Geno2pheno [coreceptor-hiv2]: a new diagnostic tool for the genotypic determination of HIV-2 coreceptor usage Döring, M*; Borrego, P; Büch, J; Martins, A; Friedrich, G; Camacho, R; Eberle, J; Kaiser, R; Lengauer, T; Taveira, N; Pfeifer, N (Saarbrücken, Germany) P352 High rates of multi-class drug resistance in HIV-1-infected individuals monitored with CD4 cell count in Uganda von Braun, A*; Scherrer, A; Sekaggya, C; Kirangwa, J; Ssemwanga, D; Kaleebu, P; Günthard, H; Kambugu, A; Castelnuovo, B; Fehr, J (Kampala, Uganda) P353 Pr evalence and impact of transmitted drug resistance in recent HIV-1 infections, Germany 2013–2015 Hauser, A*; Hofmann, A; Hanke, K; Bremer, V; Bartmeyer, B; Kücherer, C; Bannert, N (Berlin, Germany) P354 Higher rates for transmission of NNRTI-resistant viruses for subtype A versus subtype B strains in Southern Greece Kostaki, E; Sypsa, V; Nikolopoulos, G; Gargalianos, P; Xylomenos, G; Lazanas, M; Chini, M; Skoutelis, A; Papastamopoulos, V; Antoniadou, A; Papadopoulos, A; Psichogiou, M; Daikos, G; Chrysos, G; Paparizos, V; Kourkounti, S; Sambatakou, H; Sipsas, N; Lada, M; Panagopoulos, P; Maltezos, E; Hatzakis, A; Paraskevis, D* (Athens, Greece) P356 Low prevalence of pre-treatment HIV-1 drug resistance in Ugandan adults von Braun, A*; Sekaggya, C; Scherrer, A; Magambo, B; Ssemwanga, D; Kaleebu, P; Günthard, H; Kambugu, A; Fehr, J; Castelnuovo, B (Kampala, Uganda) P357 Pr evalence of resistance mutations to rilpivirine and etravirine in people starting antiretrovirals in Argentina Bissio, E*; Barbás, M; Bouzas, M; Cudolá, A; Falistocco, C; Salomón, H (Buenos Aires, Argentina) P358 Fr equency of additional resistance relevant mutations in 2% and 1% population proportions in next-generation sequencing (NGS) in routine HIV-1 resistance diagnostics Ehret, R*; Moritz, A; Schuetze, M; Obermeier, M (Berlin, Germany) P359 Impact of baseline NNRTI resistance in antiretroviral-naïve patients in a large urban clinic Steinberg, S*; Crouzat, F; Sandler, I; Varriano, B; Smith, G; Kovacs, C; Brunetta, J; Chang, B; Merkley, B; Tilley, D; Fletcher, D; Acsai, M; Knox, D; Sharma, M; Loutfy, M (Toronto, Canada) P360 Enhanced surveillance to study HIV-1 drug resistance among naïve individuals in Southern Greece: the added value of molecular epidemiology to public health Paraskevis, D*; Kostaki, E; Magiorkinis, E; Gargalianos, P; Xylomenos, G; Lazanas, M; Chini, M; Skoutelis, A; Papastamopoulos, V; Antoniadou, A; Papadopoulos, A; Psichogiou, M; Daikos, G; Zavitsanou, A; Chrysos, G; Paparizos, V; Kourkounti, S; Oikonomopoulou, M; Sambatakou, H; Sipsas, N; Lada, M; Panagopoulos, P; Maltezos, E; Drimis, S; Hatzakis, A (Athens, Greece) P361 T ransmission patterns of HIV-1 subtype A resistant strains across Greece: evidence for country and regional level transmission networks Paraskevis, D*; Skoura, L; Kostaki, E; Gargalianos, P; Xylomenos, G; Lazanas, M; Chini, M; Metallidis, S; Skoutelis, A; Papastamopoulos, V; Antoniadou, A; Papadopoulos, A; Psichogiou, M; Daikos, G; Pilalas, D; Zavitsanou, A; Chrysos, G; Paparizos, V; Kourkounti, S; Chatzidimitriou, D; Sambatakou, H; Sipsas, N; Lada, M; Panagopoulos, P; Maltezos, E; Drimis, S; Hatzakis, A (Athens, Greece) P362 Occurr ence and risk factors for primary integrase resistance-associated mutations in Austria in the years 2008–2013 Zoufaly, A*; Kraft, C; Schmidbauer, C; Puchhammer, E (Vienna, Austria) P363 T ransmission of HIV-1 drug resistance in Tel Aviv, Israel, 2010–2015 Turner, D*; Girshengorn, S; Braun, A; Tau, L; Leshno, A; Alon, D; Pupko, T; Zeldis, I; Matus, N; Gielman, S; Ahsanov, S; Schweitzer, I; Avidor, B (Tel Aviv, Israel) P364 Development of T66I-mediated integrase inhibitor cross-resistance against elvitegravir under dolutegravir-containing first-line therapy Wiesmann, F*; Däumer, M; Naeth, G; Knechten, H; Braun, P; Rump, J (Aachen, Germany) P365 Patter ns of emergent resistance-associated mutations after initiation of non-nucleoside reverse-transcriptase inhibitor-containing regimens in Taiwan: a multicenter cohort study Cheng, C*; Su, Y; Tsai, M; Yang, C; Liu, W; Cheng, S; Sun, H; Hung, C; Chang, S (Taoyuan, Taiwan) P366 Association of therapeutic failure with low-level viremia in HIV-infected patients in the Arevir/RESINA cohort in Germany Lübke, N*; Pironti, A; Knops, E; Schülter, E; Jensen, B; Oette, M; Esser, S; Lengauer, T; Kaiser, R (Düsseldorf, Germany) P367 Drug resistance mutations (DRM) among pregnant HIV-positive women in the Duesseldorf University Hospital, Germany, 2009–2016 Haars, U*; Luebke, N; Jensen, B; Haeussinger, D (Essen, Germany) P368 Pr evalence of HIV type 1 drug resistance mutations in treatment-naïve patients participating in the GARDEL study Figueroa, M*; Patterson, P; Cahn, P; Andrade-Villanueva, J; Arribas, J; Gatell, J; Lama, J; Norton, M; Sierra Madero, J; Sued, O; Rolon, M (Buenos Aires, Argentina) P369 High prevalence of transmitted antiretroviral drug resistance in newly HIV-1 diagnosed Cuban patients Perez Santos, L*; Machado, L; Kouri Cardella, V; Diaz, H; Aragones, C; Aleman, Y; Silva, E; Correa, C; Blanco de Armas, M; Perez, J; Dubed, M; Soto, Y; Ruiz, N; Limia, C; Nibot, C; Valdés, N; Ortega, M; Romay, D; Baños, Y; Rivero, B; Campos, J (Havana, Cuba) P370 V iroseq protocol optimized for the detection of HIV-1 drug mutations in patients with low viral load Monteiro, F*; Tavares, G; Ferreira, M; Amorim, A; Bastos, P; Rocha, C; Hortelão, D; Vaz, C; Koch, C; Araujo, F; Serrão, R; Sarmento, A (Porto, Portugal) P371 The role of presepsin (sCD14-ST) as an indirect marker of microbial translocation and immune activation Paola, C*; Zuccaro, V; Cima, S; Sacchi, P; Bruno, R (Pavia, Italy) P372 CRF19_cpx variant emergence in a cluster in naïve patients of southern Spain: clinical and phylogenetic characterization González-Domenech, C*; Viciana, I; Mayorga, M; Palacios, R; de la Torre, J; Jarilla, F; Castaño, M; del Arco, A; Márquez, M; Clavijo, E; Santos, J (Málaga, Spain) P373 One-step real-time PCR for HIV-2 group A and B RNA plasma viral load in LightCycler 2.0 Bastos, P; Monteiro, F*; Tavares, G; Amorim, A; Ferreira, M; Hortelão, D; Rocha, C; Vaz, C; Koch, C; Araujo, F; Serrão, R; Sarmento, A (Porto, Portugal) P374 The association between high pre-HAART CD8 cell counts and poorer immunological outcome following antiretroviral therapy Wong, C*; Wong, N; Lee, S (Hong Kong, Hong Kong) Immune recovery in acute and chronic HIV infection and the impact of thymic stromal lymphopoietin Marco Gelpi1, Hans J. Hartling1, Kristina Thorsteinsson2, Jan Gerstoft1, Henrik Ullum3, Susanne D. Nielsen1 Viro-Immunology Research Unit, Department of Infectious Diseases, University Hospital of Copenhagen Rigshospitalet, Copenhagen, Denmark 1; Department of Infectious Disease, University Hospital of Copenhagen Hvidovre, Copenhagen, Denmark 2; Department of Clinical Immunology, University Hospital of Copenhagen Rigshospitalet, Copenhagen, Denmark 3

Background Results Symptomatic primary HIV infection is associated with faster decline in Immune recovery was comparable in all groups, and no differences in CD4+ T cells count and progression to AIDS, and immediate initiation of immune homeostasis were found between primary HIV infection and combination antiretroviral therapy (cART) is recommended. However, early presenters. In primary HIV infection group, lower thymic output little is known about immunological predictors of immune recovery. compared to late presenters without advanced disease was found. Thymic Stromal Lymphopoietin (TSLP) is a cytokine that promotes However, lower proportion of effector memory and higher proportion of homeostatic polyclonal proliferation of CD4+ T cells and participates in late differentiated CD4+ T cell were found in primary HIV infection regulating Th17/regulatory T-cell balance, immunological functions known compared to late presenters. TSLP was elevated in primary HIV infection to be affected during primary HIV infection. The aim of this study was to at baseline and after 24 months of cART (Table2). Interestingly, TSLP was describe immune recovery in primary and chronic HIV infection and negatively associated with proportion of recent thymic emigrants possible impact of TSLP. (correlation coefficient -0.60, P=0.030). However, TSLP was not associated with immune recovery in primary HIV infection. Finally, higher plasma Materials and Methods TSLP was associated with lower CD4+ T cell recovery in the late presenters Prospective study including 100 HIV-infected individuals (primary HIV non advanced disease group (correlation coefficient -0.50, P = 0.034). infection (N=14), early presenters (>350 CD4+ T cells/µL, N=42), late presenters without advanced disease (200-350 CD4+ T cells/µL, N=24) and . CD4 count (A), immune recovery (B) and plasma TSLP (C) before cART and during 24 late presenters with advanced disease (<200 CD4+ T cells/μL, months of follow-up N=20))(Table1). Plasma TSLP was determined using ELISA and CD4+ T cell subpopulations (recent thymic emigrants, naïve, and memory cells) were measured using flow cytometry at baseline and after 6, 12, and 24 months of cART. Tabel 1. Clinical characteristics of the population Primary HIV (PHI) Chronic patients CD4 < Chronic patients CD4 Chronic patients HC N=14 200 (LP-AD) 200-350 (LP-nonAD) CD4 >350 (EP) N=18 N=20 N=24 N=42 Gender, males/females, (% males) 13/1 (92.9) 18/2 (90.0) 21/3 (87.5) 39/3 (92.9) 17/1 (94.4) Age, years, median (IQR) 47 (12) 42 (16) 38 (16) 44.5 (12) 42.5 (12) Time since diagnosis, days, median 2 (3) 3 (9) 18 (269) 24 (983) NA (IQR) CD4+ nadir, cells/µL, median (IQR) 540 (335) 45 (113) 290 (95) 480 (170) NA CD4+ at baseline, cells/µL, median 550 (327) 55 (110) 290 (97) 510 (172) 983 (540) (IQR) CD4/CD8 at baseline, median (IQR) 0.5 (0.3) 0.1 (0.1) 0.3 (0.1) 0.5 (0.3) 1.5 (0.9) Co-infection with chronic HBV/HCV, N 0/1 0/2 0/0 1/1 0/0 HIV-RNA at baseline, median (IQR) 151,775 (3,442,296) 196,589 (751,023) 65,990 (89,637) 49,422 (47,031) NA

AIDS defining events, N 0 1 0 0 NA Fiebig Stage I, N 1 NA NA NA NA Fiebig Stage II, N 1 NA NA NA NA

Fiebig Stage III, N 1 NA NA NA NA Fiebig Stage IV, N 11 NA NA NA NA

Tabel 2. Chronic patients CD4 < Chronic patients CD4 Chronic patients Primary HIV (PHI) 200 (LP-AD) 200-350 (LP-nonAD) CD4 >350 (EP) P* N=14 N=20 N=24 N=42 CD4 Baseline 550 (327)a,b 55.5 (110)a 290 (97)b 510 (172) < .001 Cells/µL After 24 months of cART 680 (240)a 269 (160)a 695 (290) 820 (317) < .001 RTE Baseline 14 (11) 11 (16) 20 (15) 18 (14) .063 Mann-Whitney was used to compare PHI group with the chronic groups. Significant differences are After 24 months of cART 18 (9)b 17 (10) 28 (11)b 17 (16) .009 Naive marked: a: PHI vs. late presenters with advanced disease; b: PHI vs. late presenters without Baseline 43 (20)a 23 (30)a 40 (26) 44 (21) < .001 advanced disease; c: PHI sv . early presenters After 24 months of cART 36 (12)b 30 (16) 55 (16)b 37 (15) .002 Cells EM Baseline 12 (7) 17 (14) 16 (12) 12 (6) .009 Conclusions After 24 months of cART 9 (4)a 15 (6)a 6 (7) 7 (5) .043 % CD4 CM Immune recovery was comparable in primary and chronic HIV infection Baseline 26 (6) 20 (22) 24 (10) 24 (10) .597 whereas differences in absolute counts and proportions of CD4+ T cell After 24 months of cART 24 (14) 30 (23) 24 (8) 32 (16) .784 LD subpopulations were found between primary HIV infection and late Baseline 5 (4)b 3 (12) 1 (2)b 5 (8) .009 presenters supporting early initiation of cART. Higher plasma TSLP was After 24 months of cART 9 (16)a,b 1 (1)a 3 (2)b 9 (13) .038 found in primary HIV infection. Association between TSLP and a lower TSLP /mL Baseline 2.8 (2.3)a,b,c 1.7 (0.8)a 2.1 (1.4)b 1.9 (0.7)c .023 thymic output, but not with immune recovery was found in primary HIV pg a,c a c After 24 months of cART 3.9 (2.8) 1.3(1.4) 2.4 (2.9) 2.1 (1.0) .060 infection. These findings indicate a possible role of TSLP in immune Comparing the four HIV groups by using Kruskal-Wallis test. If significant (<0.05) then Mann-Whitney was used tocompare PHI group with the other chronic groups. Only significant differences are homeostasis in HIV infection but do not support TSLP to affect immune marked: a: PHI vs LP-AD; b: PHI vs LP-nonAD; c: PHI vs EP recovery in primary HIV infection.

Contact: [email protected] P347

Cardiovascular risk in HIV positive subjects: analyses of T cell phenotype and CD49d expression Zingaropoli, Maria Antonella; Iannetta, Marco; D'Abramo, Alessandra; Oliva, Alessandra; d'Ettorre, Gabriella; Lichtner, Miriam; Mastroianni, Claudio Maria; Ciardi, Maria Rosa; Vullo, Vincenzo Department of Public Health and Infectious Diseases, Sapienza Rome Italy Background Results It is well known that HIV positive subjects have HIV positive subjects showed a lower count of CD4+ T-lymphocytes (p=0.04) and increased levels a higher risk of non-AIDS-related comorbidities of CD8+ T-lymphocytes, immuneactivation (CD4+ and CD8+ HLA-DR+CD38+, p<0.001 and than general population. Chronic p<0.001, respectively) and immunesenescence (CD4+ and CD8+ CD28-CD57+, p=0.02 and immuneactivation of T-cells plays an important p<0.001, respectively) than HD. A decrease in CD4+ and CD8+ naïve [N] (p=0.02 and p=0.01) and role in HIV pathogenesis and related an increase in CD8+ effector memory [EM] (p=0.007) percentages were observed in HIV positive comorbidities. In this context, the integrin-α4 subjects compared to HD (Figure 2). (CD49d), a transmembrane co-stimulatory molecule, is involved in the lymphocyte homing from peripheral compartment to the gut (α4β7) and to the central nervous system (α4β1). Aim of the study was to evaluate CD49d expression in T-lymphocyte subsets and the relationship with cardiovascular damage in HIV Figure 2. CD4 and CD8 naïve and CD8 effector memory percentage in HIV+ subjects compared to HD positive individuals on effective combined antiretroviral therapy (c-ART). In HIV positive subjects CD49d expression was increased on CD4+ T-lymphocyte subsets (N: p=0.01, central memory [CM]: p<0.001, EM: p<0.001, effector [E]: p=0.05) and CD8+ T-lymphocyte Materials and methods subsets (N: p=0.0006, CM: p<0.001, EM: p<0.001, E: p=0.003 and intermediate [I]: p<0.001), Thirty HIV positive subjects (6 females/24 compared to HD (Figure 3). males) with a mean age (± standard deviation [SD]) of 52±10.1 years on effective c-ART and 15 age and sex matched healthy donors (HD) were enrolled. T-lymphocyte immunophenotype and CD49d expression, (measured as median fluorescence intensity [MFI]), were assessed by flow cytometry (Figure1). Carotid-Intima Media Thickness (c-IMT) was measured with ultrasonography. Normal and pathological c- IMT were defined as IMT<0.9 mm and >0.9 mm, respectively. Figure 3. CD49d expression on CD4+ and CD8+ T-lymphocyte subsets A A positive correlation between CD49d expression in CD4+ T- cells and CD4+ HLA-DR+CD38+ (Spearman r=0.57, p=0.0012) was found in HIV positive subjects (Figure 4). In the HIV positive group c-IMT was higher (mean±SD: 0.85±0.17 versus 0.28±0.24 mm, p<0.001) than HD. CD4+ T-cell CD49d expression and CD4+HLA-DR+CD38+ were positively correlated with c-IMT (p=0.04, p=0.085, respectively) (Figure 5). Figure 4. Correlation between CD49d expression in CD4+ T-cells and CD4+ HLA-DR+CD38+

Figure 5. Evaluation of c-IMT in HIV positive patients compared HD

Among HIV positive patients, 15 (50%) had a normal c-IMT and 15 (50%) a pathological c-IMT. Moreover, HIV positive

Figure 1. Gating Strategy: T-cells immuneactivation and subjects with pathological c-IMT showed higher levels of CD4 immunesenescence were evaluated by determining the percentage CM CD49d expression (p=0.02) than HIV positive subjects of CD38 HLA-DR double positive events and the percentage of CD28- CD57+ events in the CD3+CD4+ and CD3+CD8+ gates, with normal c-IMT (Figure 6). respectively. (A) T-cell subpopulations is defined by CD45RO and CD57 markers (B). N: naïve, CM: central memory, EM: effector memory, E: effectors, I: intermediate Figure 6. Correlation between c-IMT and Conclusions CD4 CM CD49d expression In animal models a potential role of CD49d in macrophages activation has been demonstrated. In this study, the increase of CD49d expression in T- lymphocytes could be considered as a marker of immuneactivation during HIV infection. Furthermore, CD49d could represent a potential therapeutic target for the immune system modulation in the context of HIV infection aiming to reduce non-AIDS related comorbidities, especially cardiovascular diseases. Contacts: [email protected] Baseline myeloid and lymphoid ac6va6on markers can predict 6me to viral load reduc6on under 50 copies/ml and CD4 recovery, respec6vely, a?er highly ac6ve an6retroviral therapy ini6a6on

Ianne\a Marco, Lichtner Miriam, Rossi Raﬀaella, Savinelli Stefano, Vita Serena, Mascia Claudia, Zuccalà Paola, Marocco Raﬀaella, Zingaropoli Maria Antonella, mDC pDC slan-DC 100000 *** 15000 **** 150000 **** Ciardi Maria Rosa, d'E\orre Gabriella, Mastroianni Claudio Maria, Vullo Vincenzo 80000 10000 100000 P348 Sapienza University, Department of Public Health and Infec6ous Diseases, Rome, Italy 60000

40000 cells / ml / cells cells / ml / cells cells / ml / cells 5000 50000 20000

0 0 0 Table 1: Clinical characteris6c of the study popula6on Figure 5: Immunological parameters in AIDS and non-AIDS C A or B C A or B C A or B mDC pDC slan-DC ++ + presenters + ++ Background HIV+ HD CD14mDC CD16 CD14pDCCD16 100000 slan-DC*** 15000 **** 150000 **** 100000 250000 ** 100000 *** 15000 **** 80000150000 **** During HIV infec.on myeloid and lymphoid ac.va.on has 80000 200000 10000 100000 Number 34 17 80000 60000 1 60000 10000150000 100000 been reported , together with eleva.on of monocyte/ 60000 40000 cells / ml / cells ml / cells Age: median 37 37 ml / cells 5000 50000 40000 100000 40000 ml / cells ml / cells cells / ml / cells cells / ml / cells macrophage inﬂamma.on markers, such as soluble ml / cells 5000 2000050000 [IQR] [28-44] [30-49] 20000 50000 2-3 20000 0 0 0 (s)CD14 and sCD163 . The advent of highly ac.ve 0 0 C A or B C A or B C A or B Sex: M/F 26/8 13/4 0 C A or B 0 C A or B 0 an.retroviral (ARV) therapies improved both life C A or B C A or B C A or B VL: median log/ml 4,9 log/ml CD14++CD16+ CD14+CD16++ NA CD4 immuneactivation CD8 immuneactivation 4 100000 250000 expectancy and quality of life of persons living with HIV . [IQR] [4,1-5,5] 80 40 ** 60 *** 80000 200000 40 30 However, the persistence of the virus in the host leads to # CD4: median 434 cells/µl 40 NA 60000 150000 30 20 a state of chronic ac.va.on of the immune system, not [IQR] [101-656] 40000 100000 cells / ml / cells ml / cells % of CD8 % of

% of CD4 % of 20 5 CD14++CD16+ 10 CD14+CD16++ completely reversed by ARV treatment . We evaluated A1:13 A2: 5 A3: 2 10 20000 50000 100000 250000 ** 0 0 0 0 both myeloid and lymphoid ac.va.on markers and CDC classiﬁca.on B1: 1 B2: 3 B3: 1 NA 80000 C A or B 200000 C A or B C A or B C A or B correlated them with CD4 recovery aQer 12 months of C3: 9 60000 150000 CDC-C HIV+ pa#ents showed lower levels of mDC, pDC, slanDC CD8 immuneactivation 40000 100000 CD4 immuneactivation cells / ml / cells ml / cells ARV treatment 80 40 an.retroviral (ARV) treatment and the .me (in days) *** 15/10/6 NA 20000and CD4 immuneac#va#on 50000 than HD. No diﬀerences 60 in CD8 40 PI/NNRTI/INSTI 30 needed to achieve a viral load below 50 copies/ml. immuneac#va#on 0 levels 0 and intermediate monocyte 40 cell counts C A or B C A or B VL: HIV-1 viral load. HD: healthy donors. IQR: interquar#le range. were observed. 30 20 % of CD8 % of

% of CD4 % of 20 10 # CD4: CD4 absolute count. NA: not applicable. PI: protease 10 0 0 inhibitor. NNRTI: non-nucleoside reverse-transcriptase inhibitor. HIV-1 viremia nega.vely correlated with pDC and slanDC C A or B C A or B INSTI: integrase strand transfer inhibitor. cell counts and posi.vely correlated with CD14++CD16+ Matherials and Methods Mo cell counts and CD4 immune-ac.va.on levels (table Treatment-naive HIV+ pa.ents were enrolled and 2). followed up for one year aQer treatment ini.a.on. Blood At T0 HIV+ subjects showed lower levels of pDC (3.976 vs 7.043 cells/ml p<0,001) slanDC (11.644 vs 24.538 cells/ml samples were collected before treatment ini.a.on (T0). Table 2: Correla6on between HIV-1 viral load and p=0,02) and higher levels of CD14++CD16+ Mo (19.369 vs Monocyte (Mo), dendri.c cell (DC) and T lymphocyte immunological parameters at T0 phenotypes were assessed by flow-cytometry using a 7.157 cells/ml p<0,001) compared to HD (figure 2). HLA- Spearman r p lyse-no-wash protocol (ga.ng strategy is shown in Figure DR was reduced on mDC of HIV+ subjects (22.556 vs pDC -0,34 0,047 1). sCD14 and sCD163 were measured in plasma with 37.358 p<0,001) and increased on slanDC (13.680 vs 9979 ELISA. Seventeen age and sex matched healthy donors p=0,005) compared to HD. Levels of CD4+ and CD8+ T- slanDC -0,52 0,002 ++ + (HD) were enrolled. Sta.s.cal analysis was performed lymphocyte immuneac.va.on were higher in HIV+ CD14 CD16 +0,36 0,036 + + + with the soQware GraphPad Prism version 6.0. subjects than HD (6,0 vs 1,8 % p<0,001 and 9,4% vs 1,1% CD4 HLA-DR CD38 +0,50 0,002 p<0,001) (figure 3). The Kaplan-Meier analysis showed that higher baseline ++ + Figure 1: Ga6ng strategy Figure 2: pDC, slanDC and intermediate Monocytes CD14 CD16 Mo counts were predic.ve of a lower rate of subjects with a viral load <50 copies/ml, within 150 A B p DC slan-DC CD14++CD16+ 25000 150000 100000 *** * **** days from ARV therapy ini.a.on (p=0.03) (Figure 6A). 20000 80000 100000 AQer one year of ARV therapy, CD4 recovery posi.vely 15000 60000

10000

cells / ml / cells 40000 cells / ml / cells cells / ml / cells correlated with basal levels of CD8 immune-ac.va.on 50000 40 5000 20000 (Figure 6B), while the choice of trea.ng pa.ents with a PI, 0 0 0 HIV-1 HD HIV-1 HD HIV-1 HD 30NNRTI or INSTI did not aﬀect CD4 recovery. The three Comparison of pDC, slanDC and intermediate monocytes (CD14+ pa.ents who did not receive any ARV treatment were + + 20 CD16 ) Mo, between treatment naive HIV+ subjects and Healthy excluded from the analysis. Donors (HD) 10

% of CD8 HLA DR+ CD38+ DR+ CD8 HLA % of ++ + 0 Figure 6: Predic6ve value of CD14 CD16 Mo and CD8 Figure 3: T lymphocyte immuneac6va6on -1000 0 500 1000 CD4+ HLA-DR+ CD38+ CD8+ HLA-DR+ CD38+ CD4 recovery immuneac6va6on 80 **** 40 **** 60 A B 40 CD14++CD16+ Mo <16.000 30 40 ++ + 100 CD14 CD16 Mo >16.000 30 20

% of CD4 % of 20 % of CD8 % of A) Ga#ng strategy for monocytes and dendri#c cells: a4er 10 doublets exclusion, lineage (Lin:CD56, CD19, CD3, CD235a)- and 10 50 + 0 0 HLA-DR events were gated. According to CD14 and CD16 HIV-1 HD HIV-1 HD ++ - 50 VL> with % patients expression monocyte were defined as classical (CD14 CD16 ), Comparison of immuneac#va#on levels of CD4 and CD8 T 0 ++ + + ++ lymphocyte, between treatment naive HIV+ subjects and Healthy 0 50 100 150 intermediate (CD14 CD16 ) and non-classical (CD14 CD16 ). days Slan-DC were iden#fied in the CD14+CD16++ gate and defined as Donors (HD) A) Higher CD14++CD16+ Mo counts were associated to a lower rate CD11c+ and M-DC8(slan)+. Myeloid dendri#c cells (mDC) and of subjects with a viral load under 50 copies/ml, a4er ARV plasmacytoid dendri#c cells (pDC) were iden#fied in the CD14- Myeloid ac.va.on soluble markers sCD14 and sCD163 treatment ini#a#on. The cut-off of 16.000 cell/ml represents the CD16- gate and defined as HLA-DR+CD11c+ and CD11c-DC123+, were higher in HIV+ subjects compared to HD (2163 vs highest value observed in the control group. respec#vely. 1363 ng/ml p<0,001 and 272,6 vs 149,1 ng/ml p=0.085) B) CD4 recovery a4er 12 months of ARV treatment posi#vely B) Ga#ng strategy for T lymphocyte immuneac#va#on: a4er ++ + correlated with baseline CD8 immune-ac#va#on levels (Spearman + + + + (figure 4). CD14 CD16 Mo and CD8 immune-ac.va.on doublets exclusion, CD4 CD45 and CD8 CD45 lymphocytes were were not correlated with the clinical stage of HIV subjects, r: 0,50 and p:0,005). gated. Immuneac#vated CD4 and CD8 T lymphocyte were defined while pDC, mDC and slanDC cell counts were lower in as HLA-DR+CD38+. Concusions AIDS than non-AIDS presenters. CD4 immuneac.va.on pDC and slanDC are reduced in HIV+ individuals (especially in levels were higher in AIDS than non-AIDS presenter those with a CDC-C clinical stage) before ARV treatment (figure 5). Results ini.a.on. mDC are reduced in AIDS compared to non-AIDS presenters. Inflammatory CD14++CD16+ monocyte counts are We recruited 34 naive pa.ents (8 women, 9 AIDS Figure 4: Soluble inflamma6on markers presenters). 15, 10 and 6 pa.ents started an ARV therapy increased in treatment naïve HIV-1 infected pa.ents and are sCD14 sCD163 associated to a delay in viral load decrease under 50 copies/ containing a protease, a non-nucleoside reverse- 5000 **** 3000 p=0.085 transcriptase and an integrase strand transfer inhibitor 4000 ml. CD4 immuneac.va.on is associated with higher viral 2000 (PI, NNRTI, INSTI), respec.vely. Three pa.ents did not 3000 load at baseline, while higher CD8 immuneac.va.on levels ng/ml ng/ml 2000 seems to predict a higher CD4 gain, aQer 12 months of ARV start any treatment (1 elite controller and 2 long-term 1000 1000 treatment. Monocyte subsets evalua.on and lymphocyte non progressors). No differences in HIV viral load and CD4 0 0 cell counts were observed at T0, stra.fying pa.ents HIV-1 HD HIV-1 HD ac.va.on permit to easily assess .me to virological sCD14 and sCD163 levels in treatment naive HIV+ subjects and undetectability and immunological recovery according to ARV therapy. Healthy Donors (HD)

Bibliography: 1. Appay V, Kelleher AD. Curr Opin HIV AIDS. 2016 Mar;11(2):242-9. 2. Dutertre CA, Amraoui S, De Rosa A et al., Blood. 2012 Sep 13;120(11):2259-68. 3. McKibben RA, Margolick JB, Grinspoon S et al., J Infect Dis 2015 Apr 15; 211(8):1219-28. 4. Samji H, Cescon A, Hogg RS, et al., PLoS One 2013; 8:e81355. 5. Hearps AC, Maisa A, Cheng WJ et al., AIDS. 2012 Apr 24;26(7): 843-53. Contacts: marco.ianne\[email protected], [email protected] Poster number: P 349 Impact of oestrogen plasma levels in modulation of immune activation among HIV-infected women and men undergoing successful antiretroviral therapy

1 1,2 1 1,2 1 2 2 2 2 Marocco R , Lichtner M , Tieghi T , Belvisi V , Pozzetto I , Mascia C , Zuccalà P , Rossi R , Mengoni F ,

Mastroianni CM 1,2 1Infectious Diseases Department, Sapienza University, Polo Pontino, SM Goretti Hospital, Latina, Italy 2 Pubblic Health and Infectious Diseases, Sapienza University, Rome, Italy

Background:

Several sex differences have been described in the natural course of HIV-1 disease. Higher levels of TLR 7-medaited INF-aplha

production together with greater levels of activated CD8-T cells were described in women compared with men for given HIV viral load. The role of sexual hormones in ART treated women is not completely understood and seem to be crucial to individualize possible eradication strategy in women that could b different that in men. The aim of this study was to investigate the role of sexual hormones in determining innate immunity and immune activation in a cohort of HIV infected subjects undergoing effective antiretroviral treatment.

Study population Methods:

WOMEN MEN p  Whole blood samples evaluating mDC, pDC , SlanDC and

Age 50 y (24-76) 48 y(23-70) 0,33 typical, atypical and intermediate monocytes with a CD4+ Nadir 215 cell/mmc (4-640) 173 cell/mmc (8-472) 0,45 cytofluorimetric method based on 7 fluorochromes CD4+ 660 cell/mmc ( 178-1425) 709 cell/mmc (243-1550) 0,69  HLA-DR/CD38 CD4 and CD8 T lymphocytes were also HIV-RNA Zenith cp/ml 60779 cp/ml 96000 cp/ml 0,06 evaluated. HIV-RNA <20 cp/ml <20 cp/ml NS  sCD14 and sCD163 level (pg/ml) were measured by ELISA. Infection Years 16 y ( 3-27) 18 y ( 1-28 0,54  Sex hormones (oestradiol, progesterone, testosterone) Therapy: INSTI+ PI 9 6 were using CLIA kit. PI 15 14 NNRTI 17 13  Non parametric Mann-Whitney test and Spearman coefficient correlation were used. Results:

150  No significant differences in levels of 100 circulating dendritic cell (mDC, pDC) between

hormones level hormones HIV+ women and men.

0  A positive correlation was found between mDC-8 mDC-8 M W M M l e e W e ol W io e n wome n Men di d n on o on p=0,26 a a ro r r r tr tr e te te te p=0,26 s s s mDC and serum oestradiol (p=0,03, r=0,30) e es o oes o est g t g o s ro te testo p pr  A trend of increased number of atypical inflammatory monocytes and MDC-8 in %CD8+DR+: HIV+ subjects Women p=0.006 p= 0,0002 1000 r=-0.50 1000 r=-0,67 women.

%CD4+ T HLA-DR/CD38 %linfociti TCD4+ HLA-DR+ %CD8+%C T D8+ HLA-DR/CD38 HLA-DR+ 100 100 

/ml A significant augmentation of DR+38+CD4+ T /ml

15 6 pg pg

cells was found in men (p=0,02) and a 4 10 10 10 Oestradiol 2 Oestradiol negative correlation between DR+38+CD8+T 1 1 5 0 and serum oestradiol levels in all HIV subjects %CD8+(38+HLA-DR+)

% CD4+(38+HLA-DR+ ) CD4+(38+HLA-DR+ % -2 0,1 0,1 i 0 10 20 30 40 50 60 70 0 10 20 30 40 50 0 %CD8+ %CD8+in and in women was observed (respectively i Women %CD8+DR+ %CD4+ne %CD4+i n donne Menuom %CD8+DR+ + on Menm Women o D8+ D8 u C C 4+ d % % D D4 p=0,002; r-0,67; p=0,006, r=-0,50). p=0,02 C %C % p=0,90 p=0,02  Only in women a negative correlation between mDC and DR+38+CD8+ T cell was found(p=0,02; r=-0,43). sCD14 %CD8DR+: HIV+ subjects Women HIV+ subject p=0,03 6000 1e+6 p= 0.02 1000 r=0,30  Regarding soluble markers of monocytes p=0.007 1e+5 r=-0.40 r=-0,43

100 activation, we didn’t observe differences: 1e+5 4000 /ml /ml

1e+4 pg /ml

cell

cell women have a lower levels of cCD14 than 10 1e+4 pg/ml mDC

mDC 2000

Oestradiol men (pg/ml, median 2249 and 2685 pg/ml). 1e+3 1 1e+3 0 0,1 n n 1e+2 1e+2 e 0,0 2,0e+4 4,0e+4 6,0e+4 8,0e+4 1,0e+5 1,2e+5 me M 0 10 20 30 40 50 60 70 0 10 20 30 40 50 %CD8+ DR+ mDC cell/ml wo %CD8 DR+ p=0,09

Conclusions: In HIV aviremic ART treated subjects, high levels of oestrogen seem to be associated to an expansion of mDC and lower activation of CD8 T cells, underlying the importance of consider hormonal status and not only gender and age in designing immunological and therapeutic studies. geno2pheno[coreceptor-hiv2] P351 a new diagnostic tool for the genotypic determination of HIV-2 coreceptor usage

M. Doring¨ 1, P.Borrego2,J.Buch¨ 1, A. Martins2, G. Friedrich1, R. J. Camacho3, J. Eberle4, R. Kaiser5, T. Lengauer1, N. Taveira2, 6, N. Pfeifer1

1 Department for Computational Biology & Applied Algorithmics, Max Planck Institute for Informatics, Saarbrucken,¨ Germany. 2 Research Institute for Medicines (imed.ULisboa), Faculty of Pharmacy, University of Lisbon, Lisbon, Portugal. 3 Department of Microbiology & Immunology, Rega Institute for Medical Research, KU Leuven, Belgium. 4 Department of Virology, Max von Pettenkofer-Institut, Ludwig-Maximilians-University, Munich, Germany. 5 Institute for Virology, University of Cologne, Cologne, Germany. 6 Instituto Superior de Cienciasˆ da Saude´ Egas Moniz (ISCSEM), Monte de Caparica, Portugal.

Contact: [email protected] Relevance of HIV-2 coreceptor usage Results

A linear SVM (AUC=0.95) outperformed other models and was used in all subsequent analyses. For a set of 126 V3 sequences, the 10-fold nested CV sensitivity was 76.9% and the speciﬁcity was 97.3%.

All samples from a set of nine, newly phenotyped V3 sequences were classified correctly by the SVM. We validated existing markers for X4-capability [2] and identified new, significant features (p 0.05): variants 27K, 15G, and 8S. ≤ Figure 1: HIV coreceptors (www.viivhcdxresource.com) Visualization of model weights The selection of HIV-2 variants using the CXCR4 coreceptor (X4-capable) should be prevented because X4-capable variants are harder to neutralize than viruses using only CCR5 (R5)[1].

Before prescribing CCR5-coreceptor antagonists to patients infected with HIV-2, clinicans should rule out the existence of X4-capable variants. Goal: differentiate R5 and X4-capable HIV-2 variants based on the amino acid sequence of the V3 loop.

Materials and methods

Support vector machines (SVMs) were trained on a data set of 73 R5 and 52 X4-capable samples to classify binary- Figure 2: SVM weights for the V3 loop of a ROD10 isolate. encoded V3 amino acid sequences as either R5 or X4- capable. Classifier performance was evaluated using 10- fold nested cross-validation (CV). The predicted probabil- Highlights of the tool ities indicating whether a sequence originates from an X4- capable variant were transformed to false positive rates (FPRs). Accuracy: high sensitivity and specificity We developed a visual representation of position-specific Interpretability: visualization of sequence- classifier weights to indicate amino acids associated with specific weights and output of FPRs R5 and X4-capable variants (see Fig. 2). We evaluated established discriminatory sequence features from a Availability: an online web service is avail- rules-based approach by Visseaux et al. [2] and novel able at coreceptor-hiv2.geno2pheno.org features detected by the SVM using Fisher’s exact test with multiple testing correction (Benjamini and Hochberg). Opportunities: enables large-scale epidemiological studies on HIV-2 coreceptor usage References

[1] J. M. Marcelino et al. Resistance to antibody neutralization in HIV-2 infection occurs in late stage disease and is [2] B. Visseaux et al. Molecular determinants of HIV-2 R5–X4 tropism in the V3 loop: development of a new genotypic associated with X4 tropism. AIDS, 26(18):2275–2284, 2012. tool. Journal of Infectious Diseases, 205(1):111–120, 2012. P352 High rates of multi-class drug resistance in HIV-1 infected individuals monitored with CD4 cell count in Uganda

Amrei von Braun (1,2), Alexandra Scherrer (2,3), Christine Sekaggya (1), Joseph Kirangwa (4), Deogratius Ssemwanga (4), Pontiano Kaleebu (4), Huldrych Günthard (2,3), Andrew Kambugu (1), Barbara Castelnuovo (1), Jan Fehr (2)

1. Infectious Diseases Institute, College of Health Sciences, Makerere University, Kampala, Uganda 2. Division of Infectious Diseases and Hospital Epidemiology, University Hospital Zurich, University of Zurich, Zurich, Switzerland 3. Institute of Medical Virology, University of Zurich, Zurich, Switzerland 4. MRC/UVRI, Uganda Research Unit on AIDS, Entebbe Uganda

Corresponding author: [email protected]

Background Until a recent change in guidelines, HIV-infected patients on antiretroviral therapy (ART) in Uganda were monitored using CD4 cell counts only. So far, little is known about prevalence of drug resistance among HIV-infected patients with virological failure (VF) after immunological treatment monitoring in Uganda.

Methods From June 4th – September 30th, 2015, HIV-RNA was measured in HIV-infected adults (≥18 years) on ART for at least 6 months presenting to the Infectious Diseases Institute in Kampala. In case of VF (>1000 copies/mL), HIV genotyping was requested. Sequencing of partial polymerase gene was conducted using an in-house protocol. All sequences were submitted to the Stanford University HIV Drug Resistance database and the surveillance drug resistance mutations were identified using the 2009 WHO mutations list. HIV-1 subtypes were determined using REGA version 3.0.

Results HIV-RNA measurements were done in 2511 patients, who had been on ART for a median time of 4.7 years (interquartile range (IQR) 2.5-8.7). A total of 199 patients (7.9%) had VF with a median viral load of 4.4 log10 copies/mL (IQR:3.9-4.9). The majority of patients with VF (140, 70.4%) were on first-line ART, 138 (69.3%) were female, and the median age was 37 years (IQR:30-43). HIV genotyping tests were available in 163 (81.9%). HIV-1 subtypes A(46%) and D(34%) were most common. Relevant drug resistance mutations were observed in 135 (82.8%) (Figure), of which 103 (63.2%) had resistance to two drug classes, and 11 (6.8%) had resistance to all three drug classes available in Uganda.

Conclusions With 92% of all patients virologically suppressed, the overall prevalence of VF was low, and is in-line with the third of the 90-90-90 UNAIDS targets. However, the majority of failing patients had developed resistance to more than one drug class, suggesting that failing regimens – not identified as such by CD4 monitoring - had been in place for a prolonged period of time. This is a call for action to get access to close virological monitoring for patients on ART, as well as access to new antiretroviral drugs such as integrase inhibitors.

Figure: Type and frequency of most prevalent resistance-associated mutations observed. Figure legend: NRTI = Nucleoside/Nucleotide Reverse Transcriptase Inhibitors; TAM = Thymidine analogue mutation; NNRTI = Non- Nucleoside/Nucleotide Reverse Transcriptase Inhibitors; PI = Protease Inhibitors;

Acknowledgements: We would like to acknowledge all patients and their families. Funding: Swiss HIV Cohort Study, Gilead Sciences Prevalence and impact of transmitted drug resistance in recent HIV‐1 infections, Germany 2013‐2015 Andrea Hauser1, Alexandra Hofmann2, Kirsten Hanke1, Viviane Bremer2, Barbara Bartmeyer2, Claudia Kuecherer1, Norbert Bannert1 1Division of HIV and Other Retroviruses, Robert Koch Institute, Berlin, Germany ²Division of HIV/AIDS, STI and Blood‐borne Infections, Robert Koch Institute, Berlin, Germany P 353

INTRODUCTION OBJECTIVES

Transmitted drug resistance (TDR) in new HIV‐infections has  estimate the prevalence of TDR to Protease and Reverse significant clinical consequences for the treatment success. Transcriptase Inhibitors (PIs; RTIs) in new HIV‐infections among Therefore, monitoring of TDR in currently circulating HIV‐strains is newly diagnosed cases an important public health issue of the Robert Koch‐Institute.  To assess the impact on antiretroviral treatment according to the currently recommended first‐line regimens (European AIDS Figure 1: EACS Guidelines V8.0 3TC Lamivudin; ABC Abacavir; FTC Emtricitabin; TDF Tenofovir; Clinical Society (EACS) HIV Guidelines Version 8.0) Figure 1. EFV Efavirenz; RPV Rilpivirin; ATV Atazanavir; DRV Darunavir; LPV PATIENTS & METHODS Lopinavir; DTG Dolutegravir; EVG Elvitegravir, RAL Raltegravir

Sample collection Methodology Diagnostic laboratories provided dried serum  HIV‐1 genotyping was performed from “recent infections” spot (DSS) of ~60% of all newly diagnosed (<155 days: BED‐CEIA (Sedia); exclusive cases with CD4<100 HIV‐infections in Germany reported to the RKI cells/µl, CDC C) to identify resistance‐associated mutations

(2013 ‐2015). according to the WHO surveillance drug resistance mutations HIV-1 drug resistance analysis (WHO surveillance drug resistance mutation list) Linked to Information of HIV notification (Figure 2). surveillance Database RESULTS Figure 2: Workflow for sample preparation and analysis. Table 1: Characteristics of the study population  Between 2013‐2015 3,114/9,799 DSS (33%) Study population 2013‐2015 (n = 1,460) % originated from a recent infection. Of these, 1,460 Gender: male 88.1 (46%) were successfully sequenced and analyzed. female 11.4 no data 0.5  The proportion of total TDR was 10.8%, comprising Transmission group: unknown 26.0 3.8% with mono resistance to nucleotide reverse Men who have sex with men (MSM) 59.3 transcriptase inhibitors (NRTI), 2.8% to non‐NRTIs, Persons with heterosexual transmission 11.0 2.9% to protease inhibitors and 1.2% with Persons with intravenous drug use 3.6 dual/multi‐class resistances (N= 56, 41, 43, 17, respectively) Median age (IQR): 34 (27‐44) Figure 3: Proportion of HIV‐1 variants with and without (Figure 3). transmitted drug resistance. (2013‐2015; N=1,460)

Light color: low/intermediate resistance; dark color: high resistance Figure 6: Proportion of HIV‐1 variants with and without Figure 4: Number of transmitted drug resistance mutations Figure 5: Predicted susceptibility to antiretroviral drugs with transmitted resistance mutations which impact drugs from according to drug classes in the study population (N=1,460) respect to levels of resistance in the study population. first‐line regimens recommended in EACS Guidelines V 8.0. 80% (82/102) of all NRTI‐selected mutations were thymidine analogue mutations (TAMs) and T215 revertants: Considering only primary resistance mutations which M41L, K219NQR, D67EGN, T215Y, K70R, L210W and T215CDEIS, conferring low/intermediate resistance to impact drugs currently recommended in first‐line zidovudine (AZT) and stavudine (D4T). 60% (38/64) of NNRTI‐resistance was caused by K103NS conferring resistance regimens (EACS V8.0), the prevalence of TDR to efavirenz (EFV) and nevaripine (NVP). The most frequent PI‐mutations M46IL and V82FL are associated with mutations was 5.4% (0.8% NRTI; 3.1% NNRTI; 0.6% PI; low/intermediate resistance to tipranavir (TPV), nelfinavir (NFV) and fosamprenavir (FPV) (Figure 4+5). 0.9% multi drug resistance; N= 12, 45, 9, 13, respectively) (Figure 6). Conclusion TDR prevalence in recent HIV‐1 infections among newly diagnosed cases in Germany (2013‐2015) remained high (>10%) and is comparable to other European countries. TDR was mainly caused by the first‐generation NNRTI‐selected K103NS, by long‐term persisting TAMs and the PI‐selected M46IL and V82FL. While the K103NS is associated with failure of current efavirenz‐containing first‐line regimens, the impact of TAMs and frequent PI‐mutations on the success of current first‐line therapies is predicted to be low and decreases the proportion of TDR mutations relevant for initial regimens from 10.8% to 5.4%. However, to allow an optimal therapeutic sequencing, genotypic resistance testing remains important prior to treatment initiation and when switching to distinct second line regimen due to persistent mutations/T215 revertants .

Andrea Hauser HIV and other Retroviruses Robert Koch‐Institute Funding: The study was funded by the German Ministry of Health Nordufer 20 D‐13353 Berlin, Germany [email protected] Higher rates for transmission of NNRTI resistant viruses for subtype A versus subtype B P354 strains in Southern Greece

E. Kostaki1, V. Sypsa1, E. Magiorkinis1, G. Nikolopoulos2 P. Gargalianos3, G. Xylomenos3, M. Lazanas4, M. Chini4, A. Skoutelis5, V. Papastamopoulos5, A. Antoniadou6, A. Papadopoulos6, M. Psichogiou7, G.L. Daikos7, A. Zavitsanou1, G. Chrysos8, V. Paparizos9, S. Kourkounti9, H. Sambatakou10, N.V. Sipsas11, M. Lada12, P. Panagopoulos13, E. Maltezos13, S. Drimis8, A. Hatzakis1, D. Paraskevis*1

1Department of Hygiene, Epidemiology and Medical Statistics, National and Kapodistrian University of Athens, Athens, 2Medical School, University of Cyprus, Nicosia, 31st Department of Internal Medicine, G. Genimatas GH, Athens, 43rd Internal Medicine Department- Infectious Diseases, Red Cross Hospital, Athens, 55th Department of Medicine and Infectious Diseases, Evaggelismos GH, Athens, 64th Department of Medicine, Attikon GH, Medical School, National and Kapodistrian University of Athens, Athens, Laikon GH, Medical School, National and Kapodistrian University of Athens, Athens (1st Department of Medicine7 and Pathophysiology11), 8Department of Internal Medicine, Tzaneio GH, Piraeus, 9HIV/AIDS Unit, A. Syngros Hospital of Dermatology and Venereology, Athens, 10HIV Unit, 2nd Department of Internal Medicine, Hippokration GH, Medical School, National and Kapodistrian University of Athens, Athens, 122nd Department of Internal Medicine, Sismanogleion GH, Athens, 13Department of Internal Medicine, University GH, Democritus University of Thrace, Alexandroupolis *Contact Information: [email protected] .gr Introduction Results (Cont.)

We have previously found that the most prevalent  Molecular clock analyses revealed that: NNRTI resistant mutations among drug naïve 1. The time of the most recent common 4. The slope of the number of lineages individuals in Southern Greece were E138A and ancestor (tMRCA) for E138A was estimated (transmissions) over time estimated at K103N in 1992.0 [95%HPD: 1987.6-1995.6] and the exponential phase of the BDM 1982.6 [95%HPD: 1973.7-1990.6] for skylines for E138A sequences of Aim subtypes A and B, respectively (Table 2) subtype A (median: 10.13, 95%CI: 9.30- Our aim was to estimate the transmission 2. For K103N, the tMRCA was estimated in 10.90) was 10 times that of subtype B dynamics of E138A and K103N resistant strains 1999.0 [95%HPD: 1994.7-2002.5] and (median: 1.04, 95%CI: 0.96-1.11) and to investigate for potential differences 1991.8 [95%HPD: 1979.1-2000.8] for (Figure) between subtypes A and B subtypes A and B, respectively (Table 2) 5. For K103N, the slope for subtype A 3. The transmission dynamics for subtypes A transmissions was approximately 2.5 Materials and Methods and B for both E138A and K103N differed times (median: 6.16, 95%CI: 5.80-6.52) greatly (Figure) that for subtype B (median: 2.50, 95%CI: We analyzed all sequences with E138A from 179 2.45-2.55)(Figure) and 68 HIV-1 treatment naïve individuals sampled in Southern Greece during 01/01/2003 - Table 1. Distribution of transmission risk groups for the NNRTI-resistance mutations from different subtypes 31/06/2015 infected with subtype A and B, Subtype respectively. Similarly we analyzed 56 and 18 A B sequences with K103N from subtypes A and B. NNRTI-resistance mutation E138A K103N E138A K103N Sequences were available in the PT/RT Transmission risk group MSM 124 (69) 38 (68) 43 (63) 11 (61) Phylodynamic analyses were performed using a MSW 18 (10) 3 (5) 11 (16) 1 (6) Bayesian approach as implemented in PWID 9 (5) 1 (2) 4 (6) 2 (11) Other/Unknown 28 (16) 14 (25) 10 (15) 4 (22) BEASTv1.8, by using the HKY as nucleotide Total 179 (100) 56 (100) 68 (100) 18 (100) substitution model with gamma (Γ) heterogeneity MSM: Men who have Sex with Men MSW: Men who have Sex with Women PWID: People Who Inject Drugs model, an uncorrelated lognormal relaxed clock model with TipDates and the birth-death basic Table 2. Characteristics for the NNRTI-resistance mutations from different subtypes reproductive number models (BDM). Non- Subtype NNRTI-resistance mutation tMRCA (median estimate) 95% HPD Intervals informative priors were used for the MCMC runs. A The Markov chain Monte Carlo (MCMC) analysis E138A 1992.0 1987.6-1995.6 was run for each dataset for 30x10^6 generations, K103N 1999.0 1994.7-2002.5 sampled every 3.000 steps with the first 10% of B samples discarded as burn-in E138A 1982.6 1973.7-1990.6 K103N 1991.8 1979.1-2000.8 Statistical analysis for simple comparisons of the tMRCA: time of the Most Recent Common Ancestor 95% HPD Intervals: 95% Higher Posterior Density Intervals relevant distributions across different levels of categorical variables was based on Pearson’s chi- square tests as implemented in STATA 12

Results

 The distributions of transmission risk groups were similar for subtypes A and B for both E138A and K103N(Table 1)

 Specifically:

1. Men who have sex with men (MSM) Figure Bayesian skyline plots estimated by BEASTv1.8 using birth-death models (BDM) presenting the number of lineages represented 69% (N=124) and 63% (N=43) of (transmissions) over time for the NNRTI-resistance mutations (E138A, K103N) from different subtypes (A and B) infections with E138A in subtypes A and B, respectively (Table 1) Discussion 2. Similarly, MSM comprised 68% (N=38) and  This is one of the few studies highlighting  Given that the distributions of 61% (N=11) of individuals with K103N in differences in transmission dynamics of transmissions risk groups were similar subtypes A and B, respectively (Table 1) resistant strains belonging to different between the two clades (subtypes A and subtypes B), observed differences in transmission dynamics could be due to higher Acknowledgements:  Specifically, our study suggests that E138A transmissibility of subtype A or higher risk The study was in part supported by the Hellenic and K103N resistant mutations are behavior of the individuals infected with Society for the study of AIDS and STDs transmitted at higher rates in subtype A than this subtype in subtype B strains 3518131 Low prevalence of pre-treatment HIV-1 drug resistance in Ugandan adults

Amrei von Braun (1,2), Christine Sekaggya (1), Alexandra Scherrer (2,3), Brian Magambo (4), Deogratius Ssemwanga (4), Pontiano Kaleebu (4), Huldrych Günthard (2,3), Andrew Kambugu (1), Jan Fehr (2), Barbara Castelnuovo (1)

Corresponding author: [email protected]

Background Previous studies on pre-treatment drug resistance from sub-Saharan Africa have shown the highest prevalence in Uganda, particularly in Kampala, with a prevalence of 12.3%. Antiretroviral therapy (ART) has been publicly available in Uganda since 2000, with initial use - although limited - of mono/dual thymidine analogues. This study aims to describe type and frequency of pre-treatment resistance in HIV- infected Ugandan adults seeking care at one of the largest public-sector providers in Kampala, Uganda.

Methods From June 4th – September 30th, 2015, ART-naïve adults (≥18 years) presenting to the Infectious Diseases Institute (IDI) in Kampala and willing to participate in this study, were asked to give a plasma sample for pre-treatment HIV genotyping. Sequencing of partial polymerase gene was conducted using an in-house protocol. All sequences were submitted to the Stanford University HIV Drug Resistance database and the surveillance drug resistance mutations were identified using the 2009 WHO mutations list.

Results Pre-treatment drug resistance testing was available from 152 ART-naïve HIV-infected adults, of which 96 (63.2%) were female with a median age of 33 years (interquartile range (IQR) 26-41), and a median CD4 cell count of 511 cells/uL (IQR 284-713). Mutations associated with HIV drug resistance were found in 9/152 (5.9%) patients. Five patients (5/152, 3.3%) harbored NRTI mutations, and 8/152 (5.3%) had NNRTI mutations. Five (3.3%) patients had one class mutations, and 4 (2.6%) showed double class resistance. Protease inhibitor mutations were not observed (for specific mutations see table).

Drug class / mutations Total N = 152, (%)

Any NRTI mutation 5 (3.3) Conclusions K65R 1 (0.7) Contrary to previous reports, we found a low M184V 2 (1.3) prevalence of pre-treatment drug resistance among Other (M41L, T215I) 2 (1.3) Ugandan adults in Kampala. We hypothesize that the use of mono/dual thymidine analogues in the past Any NNRTI mutation 8 (5.3) contributed to a higher circulation of TAMs, as K101E 3 (2.0) observed in developed settings. The subsequent swift Y181C 2 (1.3) scale-up of triple ART in the region may have reduced K103N 2 (1.3) pre-treatment resistance over time. Other (M230L, G190A/S, Y188L) 4 (2.6)

Table: Observed transmitted drug resistance mutations

Acknowledgements: We would like to acknowledge all patients and their families. Funding: Swiss HIV Cohort Study, Gilead Sciences

Prevalence of HIV type 1 drug resistance mutations in treatment-naïve patients participating in the GARDEL Study

Maria Inés Figueroa, Patricia Patterson, Pedro Cahn, Jaime Andrade-Villanueva, José R Arribas, José M Gatell, Javier R Lama, Michael Norton, Juan Sierra Madero, Omar Sued, Maria José Rolón, on behalf of the GARDEL Study Group*

BACKGROUND MATERIALS AND METHODS

543 naïve patients from 6 countries (Argentina, Chile, Combination antiretroviral therapy has greatly reduced Spain, Mexico, Peru and US) were screened between the rate of morbidity and mortality among HIV-1 infected Dec-2010 to May 2012, and 534 HIV-sequences were patients. However, high mutation and recombination analyzed following the IAS-USA 2014 Drug Resistance rates of HIV-1 lead to the emergence of various subtypes Mutations Panel. Genotypic assays performed at and drug-resistance viruses, rendering first line ARV- screening visit were: PhenoSense HIV assay therapy ineffective in many patients. (Monogram Biosciences, San Francisco, CA, USA), The aim of this sub study is to describe the prevalence of ViroSeq HIV-1 (ViroSeq HIV-1 Genotyping System v2.0; HIV-1 subtypes and the patterns of drug resistance Celera, Alameda, CA), TRUGENE® HIV-1 Genotyping mutations among ARV-naïve HIV-1-infected patients from Assay (Siemens Healthcare Diagnostics), according to six different countries participating in the GARDEL Study availability at each site.

RESULTS

Of the 534 patients screened, 74% were Hispanic/Latino. Median time of infection at SCR was: 10.5 months. CDC HIV-1 subtypes stage A: 82%. Of 450 viral subtypes available, the most 15.9% frequent was subtype B in all three regions (Fig 1) A total SPAIN 91% of 113 samples (21.2%) had major resistant mutations; 22 US/MEX 92% samples (4.1%) had major protease mutations (M46I was 3,2% 4.1% the most common mutation: 1.5%), 85 samples (15.9 %) LA 72% had NNRTIs mutations (K103N/S was the most common INTR NNRTI IP mutation: 4.9%), and 17 samples had mutations to NRTIs B other Fig 2: Global resistance by drug class (3.2%) ,M41L (1.3%) was the most common mutation to PIs, only 2 patients had more than one mayor mutation LA US/MEX Spain (2/22)(Fig 2). The more frequent minor mutationswere:M36I/L/V(216/534),L63P(120/534), Global resistance analysis by L10I/F/V/R (115/534) and K20R/M/I:59/534. The global regions (Fig 3) 21% 22% 14% resistance analysis by regions showed 21% for LA, 22.8%

for US/Mexico and 14.7% for Spain, being NNRTI Pis* 3,1 3,1 none resistance by regions 16.4%; 15.4% and 11.8% respectively. PI resistance was 3.1% for LA and Mexico/US NNRTs 16,4 15,4 11,8 and NRTI resistance was 3.1% for LA, 3.4% for US/Mexico NRTIs 3,1 3,4 2,9 and 2.9% for Spain. No Q151M, 69ss or K65R were * major protease mutations identified.(Fig3)

CONCLUSIONS

In our study we found a primary resistance rate of 21.2%, similar in LA and US/Mexico but lower in Spain. Levels of NNRTI resistance are similar in the three analyzed regions, as previously reported in naïve populations, and reinforces the need of performing genotypic testing in ARV naïve patients, especially in LA were the first line therapy is still based on NNRTI drugs

Author correspondence: María Inés Figueroa [email protected] I�� NNRTI �� -�� SAMANTHA STEINBERG1,2, FRED CROUZAT1, INA SANDLER1, BRENDA VARRIANO1,3, GRAHAM SMITH1, COLIN KOVACS1,4, JASON BRUNETTA1, 1 1 1 1 1 1 1 1,4,5 BENNY CHANG , BARRY MERKLEY , DAVID TILLEY , DAVID FLETCHER , MEGAN ACSAI , DAVID KNOX , MALIKA SHARMA , MONA LOUTFY International Congress on Drug 14 College Street, Toronto, Ontario, Canada, M5G 1K2 Therapy in HIV Infection www.mlmedical.com 1Maple Leaf Medical Clinic; 2University of Guelph; 3Institute of Medical Science, University of Toronto; 4Department of Medicine, University of Toronto; 5Women’s College Research Institute, Women’s College Hospital 23 - 26 October 2016 Glasgow, UK

B�� R�� (��’�)

Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are particularly prone to When treated, patients without baseline NNRTI mutations (n = 1135) were prescribed NNRTI-containing regimens in treatment failure as high-level drug resistance has been associated with a single 43.9% of cases, PI-containing regimens in 34.7% of cases and INI-containing regimens in 14.3% of cases. Treated point mutations within the binding site of reverse transcriptase. Thus, transmitted patients with baseline NNRTI resistance (n = 83) were prescribed PI-containing regimens in 51.8% of cases and NNRTI mutations, may contribute to an increased risk of virologic failure for INI-containing regimens in 28.9% of cases. Baseline mutation frequencies by class are shown in Table 2. patients prescribed their first ART regimen. Virologic suppression was observed in 1024 out of 1218 (84.07%) individuals whom were prescribed ARV’s. 83.13% Our study investigated the NNRTI resistance profiles of antiretroviral–naïve of patients with baseline NNRTI mutation acheived viral suppression while 84.14% without NNRTI mutations patients in a large urban clinic setting (Maple Leaf Medical Clinic, Toronto, ON) and achieved suppression. (Table 3). assessed their response to their initial antiretroviral therapy (ART). • In univariate and mul�variate Cox regression, the presence of baseline NNRTI resistance did not impact virologic O�� suppression (HR = 0.98; 95%CI = 0.76-1.24).

The three objectives of this study are: • For virologic rebound, the presence of baseline NNRTI resistance also did not impact its occurrence (HR = 1.11; 95%CI = 0.68-1.81). 1. To assess the frequency of NNRTI, NRTI and PI mutations 2. To report if the frequency of baseline NNRTI mutations affects time to • In mul�variable analysis, a�er adjus�ng for age, gender, baseline VL and CD4 count, dura�on of HIV and baseline virologic suppression PI muta�ons, the presence of NNRTI muta�ons also did not impact virologic rebound (aHR = 1.09; 95%CI = 0.66-1.78) 3. To report if the frequency of baseline NNRTI mutation affects time to virologic (Table 4). rebound in those who have achieved virologic suppression

M��

This was a retrospec�ve clinical chart review of ART-naive pa�ents with available baseline genotypes whom were prescribed their ﬁrst ARV regimen.

Inclusion criteria:

1. HIV-posi�ve 2. Aged 16 years or older at baseline 3. Has a baseline genotype between January 1, 1997 and July 16, 2015 prior to star�ng ART

Sta�s�cal Analysis:

For demographic and clinical data, categorical variables were summarized using frequencies and propor�ons and compared using the Chi-square (Fisher) test. Con�nuous variables were summarized using medians and interquar�le range and compared using the Wilcoxon rank sum test. Baseline NNRTI, NRTI, and PI resistance muta�ons were reported (Table 2) using frequencies and propor�ons.

• Cox regression was used to determine correlates of virologic suppression [deﬁned as viral load (VL) < 40 (or <50 depending on era) by 6 months] with presence of baseline NNRTI resistance as the primary correlate.

• Of those with virologic suppression, we conducted Cox regression to determine correlates of virologic rebound (deﬁned as VL ≥ 200 copies/mL).

• Censoring occurred for those who did not have any follow-up VL results and at last VL or visit date for those without evidence of viral suppression.

R��

Baseline demographic are shown in Table 1. Of the 1338 patients with a baseline genotype, we further looked at the 1218 that subsequently initiated ARV’s.

C��

• Baseline NNRTI mutations were present in 6.7% of our antiretroviral-naive patients.

• Despite having baseline NNRTI mutations, the majority of patients (83.13%) reached virologic suppression and did not experience increased risk virologic rebound.

• Few new mutations were developed in those who started ART.

• Patients with NNRTI mutations are being treated effectively with increased use of other ARV classes. A��: Merck, Canada C��: [email protected] P 359 Enhanced surveillance to study HIV-1 drug resistance among naive individuals in Greece: P360 the added value of molecular epidemiology to public health

D. Paraskevis*1, E. Kostaki1, E. Magiorkinis1, P. Gargalianos2, G. Xylomenos2, M. Lazanas3, M. Chini3, A. Skoutelis4, V. Papastamopoulos4, A. Antoniadou5, A. Papadopoulos5, M. Psichogiou6, G.L. Daikos6, A. Zavitsanou1, G. Chrysos7, V. Paparizos8, S. Kourkounti8, H. Sambatakou9, N.V. Sipsas10, M. Lada11, P. Panagopoulos12, E. Maltezos12, S. Drimis7, A. Hatzakis1

1Department of Hygiene, Epidemiology and Medical Statistics, National and Kapodistrian University of Athens, Athens, 21st Department of Internal Medicine, G. Genimatas GH, Athens, 33rd Internal Medicine Department-Infectious Diseases, Red Cross Hospital, Athens, 45th Department of Medicine and Infectious Diseases, Evaggelismos GH, Athens, 54th Department of Medicine, Attikon GH, Medical School, National and Kapodistrian University of Athens, Athens, Laikon GH, Medical School, National and Kapodistrian University of Athens, Athens (1st Department of Medicine6 and Pathophysiology10), 7Department of Internal Medicine, Tzaneio GH, Piraeus, 8HIV/AIDS Unit, A. Syngros Hospital of Dermatology and Venereology, Athens, 9HIV Unit, 2nd Department of Internal Medicine, Hippokration GH, Medical School, National and Kapodistrian University of Athens, Athens, 112nd Department of Internal Medicine, Sismanogleion GH, Athens, 12Department of Internal Medicine, University GH, Democritus University of Thrace, Alexandroupolis

*Contact Information: [email protected] .gr Introduction Results

HIV-1 transmitted drug resistance (TDR) to NNRTIs Phylogenetic analyses revealed that: Molecular clock analyses revealed that: has been shown to compromise first-line response to  For subtype A the majority of individuals  Τhe time of the Most Recent Common treatment. The prevalence of resistance to NNRTIs was infected with resistant strains (209 out of Ancestor (tMRCA) was in 2007 (95% HPD: previously estimated to be 16.9% among drug naïve 235, 88.9%) belonged to monophyletic 2004 - 2009) for the K103N cluster versus individuals in Greece clusters (local transmission networks, LTNs) 1995 (95% HPD: 1991 - 1999), 1996 (95% (Figure 1 A). Specifically, 48 out of 56 HPD: 1989 - 2000), 1997 (95% HPD: 1991 - Aim (85.7%) of sequences with K103N, and 148 2001) and 2004 (95% HPD: 2000 - 2007) for out of 179 (82.7%) with E138A belonged to E138A LNTs (Table 3) Our aim was to investigate the dispersal patterns of one and four LNTs, respectively (Figure 1  For the K103N sub-outbreak the Re was HIV-1 resistant strains and to estimate the effective A). These findings suggest that the viruses higher than 1 between 2008 and the first half reproductive number (Re) and transmission dynamics with the most prevalent resistance of 2013 (maximum value of median Re = 2.8) for locally transmitted resistance mutations spread locally (Table 3, Figure 2). On the other hand, for all  For subtype B either non-clustered E138A LTNs the Re was higher between 2011 Materials and Methods sequences or small LTNs (range: 2-6 and 2015, except the most recent one where sequences), were identified (Figure 1 B) the Re was approximately equal to 1 (Figure We analyzed sequences from 3,428 HIV-1 treatment 2) naïve individuals available in the PT/RT. Sequences were sampled in Southern Greece during 01/01/2003 - A B 31/06/2015 Phylogenetic analysis was performed on subtype A (N=235) and B (N=86) sequences with resistance to NNRTIs (K103N and E138A) (Table 1 and 2), along with sequences isolated from seropositives without resistance from Greece sampled during 1998 - 2013 (subtype A: N=904; subtype B: N=1,615) and a randomly selected global dataset (subtype A: N=5,907; subtype B: N=3,984). Phylogenetic trees were inferred by maximum likelihood (ML) method as implemented in RAxML v8.0.20

Table 1. Distribution of HIV-1 subtypes for NNRTI-resistance mutations

NNRTI-resistance mutation (N, %) Figure 1. Unrooted ML phylogenetic trees estimated by RAxML using sequences from Greece and a global reference dataset, for HIV-1 E138A K103N subtypes: A. A and B. B. Sequences from Greece are marked in light blue in contrast with those from other geographic countries marked in dark green. Sequences with NNRTI-resistance mutations (K103N, E138A) are marked in different colors Subtype Total A 179 (68) 56 (70) 235 (69) B 68 (26) 18 (23) 86 (25) Other 16 (6) 6 (7) 22 (6) Total 263 (100) 80 (100) 343(100)

Table 2. Distribution of transmission risk groups and sampling periods for the NNRTI-resistance mutations from different subtypes Subtype A B

NNRTI-resistance mutation E138A K103N E138A K103N Sampling period 2003-2015 2004-2015 2003-2015 2004-2014 Transmission risk group MSM 124 (69) 38 (68) 43 (63) 11 (61) MSW 18 (10) 3 (5) 11 (16) 1 (6) PWID 9 (5) 1 (2) 4 (6) 2 (11) Other/Unknown 28 (16) 14 (25) 10 (15) 4 (22) Total 179 (100) 56 (100) 68 (100) 18 (100) MSM: Men who have Sex with Men MSW: Men who have Sex with Women PWID: People Who Inject Drugs

Phylodynamic analyses were performed using birth- Figure 2. Bayesian skyline plots estimated by BEAST2 using birth-death models (BDM) presenting the effective reproductive number death models (BDM) allowing estimation of important (Re) over time for the five transmission networks epidemiological parameters such as the effective reproductive number (Re), as implemented in BEAST2. Discussion The Re is defined as the number of expected secondary infections per infected individual  Our study suggests that the most prevalent mutations associated with resistance to NNRTIs were transmitted through local networks in Greece Table 3. Characteristics for the five transmission networks  Notably, phylodynamic analysis allows estimating that resistance in the last few years has been

Transmission tMRCA Re actively propagated with an increasing incidence network (median; 95% HPD) (maximum value of median)  Those belonging to the active TDR networks are the priority population for prevention (TasP) K103N 2007 (2004-2009) 2.8  Our study highlights the added value of the latest advances in molecular epidemiology to public E138A_1 1995 (1991-1999) 2.1 health since these allow us to estimate critical epidemiological parameters and therefore the E138A_2 1996 (1989-2000) 1.8 E138A_3 1997 (1991-2001) 2.0 priority population to intervene E138A_4 2004 (2000-2007) 2.5 tMRCA: time of the Most Recent Common Ancestor Re: Effective reproductive number Funding: This study has been supported by Gilead Sciences Transmission patterns of HIV-1 subtype A resistant strains across Greece: P361 Evidence for country and regional level transmission networks

D. Paraskevis1, L. Skoura2, E. Kostaki1, E. Magiorkinis1, P. Gargalianos3, G. Xylomenos3, M. Lazanas4, M. Chini4, S. Metallidis2, A. Skoutelis5, V. Papastamopoulos5, A. Antoniadou6, A. Papadopoulos6, M. Psichogiou7, G.L. Daikos7, D. Pilalas8, Α. Zavitsanou1, G. Chrysos9, V. Paparizos10, S. Kourkounti10, D. Chatzidimitriou11, H. Sambatakou12, N.V. Sipsas13, M. Lada14, P. Panagopoulos15, E. Maltezos15, S. Drimis9, A. Hatzakis1

1Department of Hygiene, Epidemiology and Medical Statistics, National and Kapodistrian University of Athens, Athens, 2Department of Microbiology, AHEPA University Hospital, Aristotle University of Thessaloniki, Thessaloniki, 31st Department of Internal Medicine, G. Genimatas GH, Athens, 43rd Internal Medicine Department-Infectious Diseases, Red Cross Hospital, Athens, 55th Department of Medicine and Infectious Diseases, Evaggelismos GH, Athens, 64th Department of Medicine, Attikon GH, Medical School, National and Kapodistrian University of Athens, Athens, Laikon GH, Medical School, National and Kapodistrian University of Athens, Athens (1st Department of Medicine7 and Pathophysiology13), 8Medical School, Aristotle University of Thessaloniki, Thessaloniki, 9Department of Internal Medicine, Tzaneio GH, Piraeus, 10HIV/AIDS Unit, A. Syngros Hospital of Dermatology and Venereology, Athens, 11Department of Microbiology, Μedical School, Aristotle University of Thessaloniki, Thessaloniki, 12HIV Unit, 2nd Department of Internal Medicine, Hippokration GH, Medical School, National and Kapodistrian University of Athens, Athens, 142nd Department of Internal Medicine, Sismanogleion GH, Athens, 15Department of Internal Medicine, University GH, Democritus University of Thrace, Alexandroupolis

Introduction Materials and Methods

The prevalence of mutations conferring resistance We analyzed sample of subtype A1 Phylogenetic topology (tree) was to NNRTIs was previously reported to be higher sequences (N=1,104) available in the pol estimated from the underlying nucleotide than 15% among drug naïve individuals both in gene (PT/RT) sequences using approximate maximum Northern and Southern Greece. The most prevalent likelihood (ML) method with bootstrapping resistance mutations were E138A, K103N and Sequences were sampled in Northern and as implemented in RAxML v8.0.20 Y181C associated mostly with subtype A1 Southern Greece during 1999 and middle- Specifically, analysis was performed under 2015. We included sequences only from the Generalized Time Reversible Greece since we have shown previously (GTR+cat) model of nucleotide Our aim was to investigate the dispersal patterns of that subtype A1 sequences have been substitution model including a Γ distributed HIV-1 resistant strains across Greece mostly found within a single monophyletic rate of heterogeneity among sites cluster

Results

Phylogenetic analysis revealed that:

 E138A and K103N resistant strains have spread through large monophyletic clusters spanning both Northern and Southern Greece, suggesting that all transmissions within these clusters occurred regionally (Figure)  Conversely to E138A and K103N, Y181C formed a subnetwork (monophyletic cluster) limited in Northern Greece with only a single spill over to Southern Greece (Figure)  For K103N strains we found a large (N=49) and a small cluster (N=5) including sequences from both areas (Figure)  Sequences from Northern Greece formed two specific subnetworks, suggesting local dispersal (Figure)  Sequences with E138A from Northern Greece formed two specific subnetworks within the E138A monophyletic clades found for Greece. The latter consisted of four major clades of 53, 41, 29 and 25 sequences from both regions (Figure)  Overall, E138A and K103N spread through common networks across the country with evidence of local transmissions in Northern Greece (Figure) Figure. Unrooted ML phylogenetic tree estimated by RAxML using HIV-1 sequences from Greece. Sequences without NNRTI resistance mutations from Northern and Southern Greece are marked in  On the other hand, Y181C has spread only in Northern Greece with light purple in contrast with those with NNRTI resistance mutations (E138A, K103N, Y181C) marked in different colors. Sequences from Southern Greece with NNRTI resistance mutations are shown in red very limited dispersal to Southern Greece (Figure) (E138A) and yellow (K103N). Sequences from Northern Greece with NNRTI resistance mutations are shown in blue (E138A), green (K103N) and light blue (Y181C)

Discussion

 A high prevalence of NNRTI resistance mutations was previously  Significant clustering of sequences from Northern Greece as well as reported for the subtype A1 strains circulating in Greece and the existence of a regional cluster suggest high transmission especially in Northern Greece networking of the population in this area; a finding that might explain the higher prevalence of transmitted drug resistance (TDR) in  The majority of these resistant viruses were transmitted within Northern Greece common transmission networks  Our study highlights the priority population to prevent TDR in the future

Acknowledgments: The study was in part supported by the Hellenic Society for the study of Contact Information: [email protected] AIDS and STDs Occurrence and Risk Factors for Primary Integrase Resistance- associated Mutations in Austria in the years 2008-2013 A. Zoufaly 1, Kraft C1, Schmidbauer C1, Puchhammer-Stöckl E2

1 Department of Medicine IV, Kaiser Franz Josef Hospital, Vienna, Austria; 2Department of Virology, Medical University Vienna, Austria

Introduction: Methods: In Europe, country specific treatment guidelines often do not advocate testing Samples of ART naïve patients in Austria between 2008 and 2013 were for Integrase inhibitor resistance associated mutations (IRAM) before initiation analyzed for the existence of IRAM using bulk sequencing with published of first line ART given the extremely low prevalence of mutations found in primers and drug resistance penalty scores (Stanford HIVdb algorithm) were older surveillance studies. However, increased use of integrase inhibitors calculated to estimate response to antiretroviral drugs Demographic and (INSTI) might have led to the emergence of treatment limiting mutations in virological data including age, sex, viral subtype, drug resistance associated more recent years. We aimed to determine the prevalence of IRAM in Austria mutations to PI and RTI were extracted from a database. Comparative statistics in the 5 years following introduction of INSTI and to analyze trends and factors and logistic regression models were used to analyse risk factors for the associated with their detection. occurrence of IRAM.

Variable Dolutegravir year Raltegravir (n) Elvitegravir (n) Mean age (years, SD) 38 (12) (n) Male sex (n, %) 235 (77.6%) 0-9 10-14 15-29 30-59 0-9 10-14 15-29 30-59 0-9 10-14 Year of sample (n, %) 2008 47 0 2 0 47 1 1 0 49 0 2008 49 (16.2%) 2009 47 0 2 2 47 0 2 2 50 1 2009 51 (16.8%) 2010 50 0 1 0 50 0 1 0 51 0 2010 51 (16.8%) 2011 48 0 2 0 48 0 2 0 50 0 2011 50 (16.5%) 2012 46 1 5 1 46 3 3 1 53 0 2012 53 (17.5%) 2013 48 0 1 0 48 0 1 0 48 1 2013 49 (16.2%)

Viral load categories (HIV RNA copies/ml,n,%) 0-9 susceptible 10-14 potential low level 15-29 low level resistance 30-59 intermediate level resistance <1x104 7 (6.7%) 1x104<5x104 23 (21.9%) Table 2: Drug penalty scores (Stanford HIVdb algorithm) indicating susceptibility to Integrase inhibitors 5x104<1x105 20(19.1%) 1x105<1x106 37(35.2%) >1x106 18(17.1%) Viral subtype (n, %) Patient Sex Age Subtype Year of IRAM RTI resistance PI A 22 (7.3%) detection mutation B 188 (62.1%) 1 male 45 B 2008 T97Aa none none C 35 (11.6%) 2 male 30 B 2009 G140Aa none A71Vc other 58 (19.1%) 3 female 31 CRF01_AE 2012 T97Aa none none Major/primary drug resistance present (n, %) 4 female 22 D 2012 L74Ma none none NRTI 4 (1.3%) 5 male 30 B 2012 F121Yb none none NNRTI 20 (6.6%) 6 male 48 C 2012 T97Aa none none PI 5 (1.7%) 7 male 39 B 2013 E138Ka none none INSTI 1 (0.3%)

Table 1: Patient Characteristics Table 3: Patients with Integrase resistance associated mutations

Results: A total of 303 samples were analyzed. Patient characteristics are shown in Table 1. Risk factor OR 95% CI P Overall prevalence of IRAM was 2.3%. Male sex 0.80 0.36 27395 0.58 6% had a DPS >=10 for Raltegravir or Elvitegravir, respectively, indicating at least Age (per year) 0.98 0.96 1.01 0.26 potential low level resistance. 1% had a GSS >=10 for Dolutegravir (Table 2). PI/NRTI or NNRTI mutation 0.44 0.10 1.96 0.28 One major mutation was observed (F121Y) in a patient sample from 2012 leading Calendar year 1.04 0.86 1.27 0.44 to 5-10 fold reduced susceptibility to Raltegravir and Elvitegravir Two patients Subtype B virus 0.54 0.06 4.60 0.57 carried the major accessory mutations E138K and G140A, respectively, which both lie on the Q148 pathway (Table 3). No temporal trend was observed (p=0.16). Risk factors associated with occurrence with IRAM are shown in Table 4. Table 4: Risk factors for Integrase resistance associated mutations

Conclusions:  Major primary IRAM are rarely found despite increasing use of INSTI in Austria  There is potential for reduced susceptibility to these drugs in selected patients  No clear risk factors for occurrence of IRAM can be identified  Routine resistance testing seems prudent to avoid the consequences including accumulation of further mutations and therapeutic failure

References Acknowledgement and financial disclosure 1) Stekler JD et al, Antivir Ther. 2015;20(1):77-80 The performance of this study was supported by GILEAD 2) Saladini F et al, Clin Microbiol Infect. 2012;18(10) 3) Gutierrez C et al, HIV Clin Trials. 2013;14(1):10-6 4) DAIG, Deutsch-Österreichische Leitlinien. 2015 . Transmission of HIV-1 Drug Resistance in Tel-Aviv, Israel, 2010-2015 Dan Turner1*, Shirley Girshengorn1,3, Adi Braun1, Luba Tau1, Ari Leshno1, Danny Alon1, Tal Pupko2, Irena Zeldis1,3, Svetlana Ahsanov1,3, Simona Gielman1,3, Natasha Matus1,3, Inbal Schweitzer1,3 and Boaz Avidor1,3 1Crusaid Kobler AIDS Center Tel-Aviv Sourasky Medical Center, affiliated to the Sackler Faculty of Medicine , Tel-Aviv University, Tel-Aviv, Israel 2Department of Cell Research and Immunology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Ramat Aviv 69978, Israel. 3Laboratory for Viruses and Molecular Biology, Tel-Aviv Sourasky Medical Center, Tel-Aviv, Israel.

Background: As of 2010, testing for HIV drug Table 1. Demographic Table 1b. Percentage of HIV subtypes among Characteristics different exposure risk category (ERC) resistance is performed routinely to all new HIV Age mean patients followed-up in Tel-Aviv. Thus, the objective Subtypes A B C other 36.8 (sd 10.1) ERC n (%) of this study was to evaluate transmission drug Gender n (%) MSM 27 7.0 320 83.3 10 2.6 27 7.0 IVU 83 86.5 11 11.5 0 0.0 2 2.1 resistance mutations (TDR) among HIV-1 Male 545 81.5 Hetero 71 41.0 37 21.4 42 24.3 23 13.3 Female 124 treatment-naïve patients in Tel Aviv from 2010 to 18.5 Unknown 5 55.6 3 33.3 1 11.1 0.0 2015. ERC Child 3 42.9 1 14.3 0.0 3 42.9 MSM 384 57.4 MSM - Men who have sex with men. IVU - Intravenous drug users. Hetero - Heterosexuals. IVU 96 14.3 Methods: The first blood samples obtained Heterosexuals 173 25.9 between 2010 and 2015 from each treatment-naïve Unknown 9 1.3 Table 2. Transmission of HIV drug resistance-associated patients joining Tel Aviv HIV clinic after the Child 7 1.0 mutations* diagnosis of HIV were sequenced for protease and Country TDR by Drug Classes origin reverse transcriptase (RT) regions. TDR in these Year n NRTI NNRTI PI MDR TDR % Israel 363 54.3 2010 119 4 6 7 1 18 15.1 two regions were defined according to the criteria Ex-Soviet Union 216 32.3 2011 100 1 3 6 3 13 13.0 proposed by Bennett et al. [2009. PLoS One. Sub Saharan Africa 42 6.3 2012 112 3 3 1 1 8 7.1 2013 133 4 3 1 8 6.0 Western Europe 12 1.8 4:e4724].Subtyping of the isolates was based on 2014 101 5 7 2 14 13.9 South¢ral the Stanford HIV Drug Resistance Database. 2015 104 5 7 3 15 14.4 America 17 2.5 Phylogenetic reconstruction was inferred using pol Total 669 18 30 22 6 76 11.4 North America 4 0.6 *Excluding the integrase inhibitor region sequences. Multiple sequence alignments were East Asia 3 0.4 TDR, transmission drug resistance mutation; NRTI, nucleoside reverse transcriptase inhibitor; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor computed using MAFFT v. 7. The phylogenetic tree EMEA 2 0.3 was then inferred using maximum likelihood as North Africa 4 0.6 implemented in PHYLIP. Likelihood computations Australia 2 0.3 Unknown 4 0.6 Table 3. Rate of transmission of drug resistance were based on the HKY model, taking among site MSM - Men who have sex with men. mutations by exposure risk category (ERC)* rate variation into account (i.e., the gamma IVU - Intravenous drug users. MSM Heterosexuals IVU EMEA - Eastern Mediterranean European and Year n TDR % n TDR % n TDR % Middle East countries, except Israel and North Africa. parameter). Confidence in tree estimation was 2010 68 13 19.1 34 5 14.7 15 0 0.0 2011 63 8 12.7 20 2 11.8 16 3 18.8 based on 100 bootstrap replications. Ethical 2012 66 6 9.1 23 1 4.3 22 1 4.5 2013 77 7 9.1 30 1 3.2 22 0 0.0 approval for the study was granted by the 2014 63 11 17.5 26 2 7.7 9 1 11.1 institutional ethics committee. 2015 47 8 17.0 40 5 14.3 12 0 0.0 MSM - Men who have sex with men. IVU - Intravenous drug users. Results: Table 1 shows the characteristic of the patients. MSM was the major exposure risk category (ERC) group followed in Tel-Aviv, 76 % among them were born in Israel, and 83 % harbor subtype B viruses. Other groups include intravenous drug users (IVU); 78 % of them were Phylogenetic tree analysis of 610 HIV-1viruses from subtypes A, B and C. born in the former Soviet Union countries and 86% Sequences of patients with TDR are represented by colored branches. harbor subtypes A viruses. The heterosexuals group is very heterogeneous and includes patients born in Israel, Ethiopian immigrants, immigrants from the former Soviet Union, and worker immigrants mainly from Africa. Rate of TDR is described in tables 2 and 3. The resistance rate decreased from 15.1% in 2010 to 6% in 2013 ( P < 0.05). In 2014 and 2015 we noted an increase to 13.9% and 14.4% respectively. In 2010-2011 protease inhibitors (PIs) was the major resistance mutation , while in 2014-2015 NNRTI resistance mutation was dominant.

Phylogenetic analysis of subtypes A, B and C was Figure A : 188 Sequences. Outgroup represented by 2 subtype B viruses. Figure B : 370 Sequences. Outgroup represented by 2 subtype C viruses. performed on 610 sequences (Fig. 1). In subtype A viruses we found a cluster among IVU at 2012 during an outbreak, without resistance associated mutation. However, a cluster with viruses harboring resistance mutation at position 103 was found in five MSM and one IVU female. The analysis subtype B viruses support clustered TDR among MSM. Among subtype C viruses there were no specific clusters.

Discussion: TDRs among patients followed in Tel- Aviv were represented by clusters in MSM. These clusters contain resistance associated mutations to drugs less prescribed in recent years. Although the region of the integrase gene is not routinely Figure C : 52 Sequences. Outgroup represented by 2 subtype viruses. sequenced in treatment-naïve patients followed-up These trees support clustered transmission of TDR throughout this period among MSM in subtype A and B. One cluster in subtype A without TDR represents an outbreak. in our center, low rate of InI TDR is reported in other Lack of clusters among IVU Harboring Subtypes A and C could represent infections acquired Before immigration to Israel from former Soviet Union countries and Ethiopia, studies. respectively. Poster-No.: P364

Development of T66I-mediated integrase inhibitor cross-resistance against elvitegravir under dolutegravir containing firstline therapy

Wiesmann F.1, Braun P.1, Naeth G.1, Rump JA2 and Heribert Knechten1 1 PZB, Aachen, HIV&Hepatitis Research Group, Aachen, Germany 2 Medical Center for Internal Medicine and Rheumatology, Freiburg, Germany

Background Results (continued) As second generation integrase inhibitor (INI), dolutegravir 50 (DTG) has shown a superior barrier to resistance as 45 compared to profiles of raltegravir (RAL) or elvitegravir 40 (EVG). Current findings suggest that resistance mutations 35 30 against INIs (Fig.1) extreme rarely occur under DTG-

% 25 containing first line antiretroviral therapy (ART)*. This case 20 report unveils a possible development of a T66I-mediated 15 cross-resistance against EVG under a DTG firstline regimen 10 5 (Tab.1). N= 0 0 0 1 4 1 1 0 1 3 1 0 2 2 2 3 2 5 5 3 2 5 1 3 2 5 5 3 3 0 4 5 2 3 2 3 121111 8 1 0 0 2 1 1 1 3 4 2 0 Methods T66I/A L74M E92Q T97A G140S Y143C/R Q148H N155H E157Q G163R/K 2009-2010 2011-2012 2013 2014 2015

A firstline treatment with lamivudin/abacavir, lopinavir and Fig.1: Prevalence of integrase-mutations in patients with confirmed integrase-inhibitor resistance dolutegravir was initiated by a 44 years old man with a 2009 – 2015 in our centre (Patients with INI-RAMs: 2009-2010; n=15 / 2011-2012; n=25 / 2013; n=11 diagnosis of HIV in 11/2015 (CDC status: B2, CD4 nadir: / 2014; n=23 / 2015; n=14) 219/µl, HIV-1 RNA: 350,000 copies/mL). Ultra-deep Tab.1: Prevalence of resistance-associated substitutions at position T66 in HIV-1 integrase sequencing was performed by using population sequencing Year of analysis T66 variant RT/P ART Subtype and ultra-deep sequencing (UDS, Illumina MiSeq) at resistance? baseline and at time of therapy failure. Resistance 2014 T66A Yes LPV/r, RAL B interpretation was estimated by using the HIV-Grade 2015 T66A No Naive B 12/2015, Stanford HIV-db 7.0.1, Rega 9.1.0 and the ANRS T66K Yes TDF/FTC/COB/ B 25_09/2015 database. Viral load was quantified with Abbott EVG Realtime. T66I Yes TDF/FTC, CRF02_AG DRV/r Results T66I Yes TDF/FTC/COB/ B EVG

Before start of therapy, no resistance-associated variants 2015/16 T66I No 3TC/ABC, B could be detected neither by population or by UDS in HIV LPV/r, DTG protease, reverse transcriptase and integrase (Table 2). After start of DTG-firstline therapy, HIV viral load dropped from Tab.2: Case report: Development of the T66I variant under DTG-containing firstline-treatment. 302,815 copies/ml to 2,400 copies/ml within four weeks of Date 11/2015 12/2015 01/2016 04/2016 05/2016 06/2016 follow up and was undetectable at week 8. CD4 cell counts ART 3TC/ABC, 3TC/ABC, 3TC/ABC, 3TC/ABC, 3TC/ABC, 3TC/ABC, increased from 219/µl to 479/µl (13.4%). However, 20 LPV/r, DTG LPV/r, DTG LPV/r, DTG LPV/r, DTG LPV/r, DTG LPV/r, DTG weeks after initiation of ART, HIV viral load increased to DTG TDM --- 1599 ng/ml ------2619 ng/ml --- (opt. (18 h) (10 h) 105 copies/ml and maintained low viremic four weeks later >500ng/ml) at 112 copies/ml most likely due to inadequate adherence Viral load 303,815 2,430 0 105 112 0 although plasma drug levels turned out to be above critical CD4 (abs.) 219 313 479 427 485 587 limits. Resistance No RAMs ------T66I/T --- (Pop.-Seq) More importantly, the development of the INI-resistance Resistance No RAMs ------T66I --- mutation T66I was then verified by UDS showing a (UDS) (36,1%) minority population of 36.1%. The variant T66I is a non- polymorphic mutation and reduces EVG susceptibility by Conclusions ~15-fold while susceptibility to RAL or DTG is reported to be unaffected. There was no evidence for protease or reverse Although being extreme rarely observed, INI-resistant HIV transcriptase resistance mutations at this time. 28 weeks variants may also occur under DTG firstline treatment. The T66I after start of therapy the viral load decreased to undetectable alone does not necessarily limit the susceptibility to DTG itself levels without any changes. but could be a first step of resistance development against DTG. Corresponding adress: PZB Aachen / Blondelstr. 9 / 52062 Aachen / Germany / Phone: +49-241-470970 / E-mail: [email protected] It is reported that T66I confer high-level resistance against EVG

*Lepik KJ et al. 2016 – CROI 2016, poster 492LB and may also putatively lower the resistance barrier against RAL. Abstract No. P365 Patterns of emergent resistance-associated mutations after initiation of non-nucleoside reverse-transcriptase inhibitor-containing regimens in Taiwan: a multicenter cohort study Chien-Yu Cheng1 , Yi-Ching Su2, Wen-Chun Liu2, Shu-Hsing Cheng1, Hsin-Yun Sun2, Chien-Ching Hung2, Sui-Yuan Chang3 Correspondence: Chien-Yu Cheng 1Department of Internal Medicine, Taoyuan General Hospital, Ministry of Health and Welfare, Taoyuan, Taiwan E-mail: 2Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan [email protected] 3Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan

Background Results Non-nucleoside reverse-transcriptase inhibitor (NNRTI)-containing During the 3.5-year study period, 1642 patients initiated nNRTI-containing antiretroviral therapy (ART) remains the recommended first-line regimens regimens, and 454 (27.4%) had to switch first-line ART because of adverse effects for adults infected with HIV in many resource-limited countries. Increasing or intolerance (n=323, 19.7%), retrospective detection of RAMs at baseline (41, trends of resistance-associated mutations (RAMs) to nNRTIs have caused 2.5%), and virological failure (83, 5.1%). Virological failure to 2 NRTIs plus concerns about the effectiveness of the regimens in national programs in nevirapine, efavirenz, and rilpivirine with baseline PVL < 5 long10 was 4.9% these regions [1-3] . In this multicenter study, we aimed to investigate the (12/245), 1.9% (11/573), and 0.7% (2/277); virological falure to 2 NRTIs plus incidence of emergent RAMs of HIV-1 to antiretrovirals (ARVs) in HIV- nevirapine and efavirenz with baseline PVL > 5 log10 was 16.4% (29/177) and positive adults who developed virological failure to first-line nNRTI- 7.8% (29/373) respectively (Figure 1). In 68 patients (3.8%) emergent RAMs were containing ART in Taiwan. identified: 42 patients (62.7%) with NRTI RAMs; 28 (41.2%), 1 (1.5%) and 48 Methods patients (71.6%) with nNRTI, protease inhibitors (PI), and any ARV RAMs, Between June 2012 and March 2016, ARV-naïve HIV-positive adults who respectively, and 21 (31.3%) with resistance to 2 or more classes of ARV. The initiated 2 NRTIs plus NNRTI at participating hospitals were included for common emergent RAMs to NRTIs were K65R(25%), M184I(10.3%), and M184V analysis. Plasma HIV RNA load (PVL) was determined at baseline, and (36.8%), and RAMs to nNRTIs included V90I (5.9%), K101E (5.9%), K103N week 4-6 and subsequently every 12 to 16 weeks after ART initiation. (19.1%), V108I (7.4%), Y181C(11.8%), and G190A (5.9%) (Figure 2).

Virological failure was defined as a decrease of PVL <1.0 log10 copies/ml in Figure 2. Patterns of emergent resistance-associated mutations after initiation 4 to 6 weeks of ART initiation; or PVL ≥200 copies/ml at 6 months of ART of three different non-nucleoside reverse transcriptase inhibitor-containing initiation; or confirmed HIV RNA ≥ 200 copies/ml after viral suppression regimens. (PVL<50 copies/ml). Population sequencing was used to detect RAMs. RAM of NNRTI (%) 6.00% Detection of RAMs at baseline was performed retrospectively. RAMs were interpreted using the IAS-USA 2015 mutations list. Figure 1. Prevalence of emergent resistance-associated mutations 5.00% among three different non-nucleoside reverse transcriptase inhibitor- containing regimens. 4.00%

18.0% P=0.003 3.00% 16.0% 16.4%

14.0% 2.00% P=0.024 12.0% 1.00% P=0.004 10.0%

0.00% 8.0% 7.8% NVP EFV RPV V90I A98G L100I K101E 6.0% K101P K103N K103S V108I 4.9% E138G V179D + K103R V179DEV V179E 4.0% P=0.513 V179T Y181C Y188L G190A G190Q G190S H221HY P225H 2.0% 1.9% M230L 0.7% Conclusions 0.0% > 5 log10 < 5 log10 > 5 log10 < 5 log10 < 5 log10 While a substantial proportion of the patients discontinued first-line NNRTI- NRTIs + NVP NRTIs + EFV NRTIs + RPV containing regimens due to adverse effects, virological response to nNRTI- containing regimens remained good in patients who were able to tolerate the References 1. Rhee SY, et al. PLoS Med 2015; 12:e1001810. regimens in Taiwan. Most common RAMs in those with virological failure were 2. Lai CC, et al. J Antimicrob Chemother. 2012;67(5):1254-60. related to exposure to tenofovir disoproxil fumarate, lamivudine, nevirapine, and 3. The TenoRes Study Group. Lancet Infect Dis 2016;16(5):565-75. efavirenz. Association of therapeutic failure to low level viremia in HIV-1 infected patients in the AREVIR/RESINA cohort in Germany Nadine Lübke1, Alejandro Pironti2, Elena Knops3, Björn Jensen4, Mark Oette5, Stefan Esser6, Thomas Lengauer2, Jörg Timm1 and Rolf Kaiser3 for the Resina Study Group 1 Institute of Virology, University of Düsseldorf, Germany; 2 The Computational Biology and Applied Algorithmics Department, Max Planck Institute for Informatics, Saarbrücken, Germany; 3 Institute of Virology, University of Cologne, Germany; 4 Department of Gastroenterology, Hepatology and Infectiology, University of Düsseldorf, Germany ; 5 Clinic for General Medicine, Gastroenterology and Infectious Diseases, Augustinerinnen Hospital , Cologne, Germany; 6 Department of Dermatology, University of Duisburg Essen, Essen, Germany

BACKGROUND § LLV has been previously associated to virological failure (VF) [1, 2] § therapeutic success: reduction of the HIV-1 viral load (VL) below 50 copies/ml (German-Austrian guidelines for the treatment of HIV infection) § Low level viremia (LLV): repeated VL measurements between 50 and 200 copies/ml after initial therapeutic success

OBJECTIVES RESULTS II § independent analysis of the association of LLV and other factors with § Most risk of low level viremia (Figure 2): VF § PI-based therapies: 165/294 (56.1%) MATERIAL & METHODS § NNRTI-based therapies: 76/294 (25.9%) • AREVIR/RESINA database: clinical and virological data of therapy-naïve § Comparable VF rates of NRTI-, NNRTI- and PI-based therapies (Ø=20%, (TN) and therapy-experienced (TE) HIV-1-infected patients in North range 17.1-22.2%) (Figure 2) Rhine-Westphalia, Germany § VF was never related to entry inhibitors or integrase inhibitors • Query of the database: • 2,485 first line and 3657 further-line therapies § No risk of VF subsequent to LLV with drugs approved ≥ 2005 (p<0.001) • patients who attained confirmed therapeutic success under ART (Figures 3 and 4) and experienced confirmed LLV thereafter • therapies in which the VL was measured at least once every 24 weeks • VF: confirmed viral load greater than 200 copies/ml following therapeutic success • p-values were calculated with Fishers’ exact and Wilcoxon rank sum test.

RESULTS I § LLV occurred in 294/6142 documented therapies (4.8%) (Figure 1) § First-line: 47/2485 (1.9%) § Further-line: 247/3657 (6.8%) § Mean time to LLV: 27 months (σ=20.7) Figure 22: Risk of virological failure after low level viremia according to the drug class. Virologicalfailure rates present the percentages of the drug related low level viremia prevalence. § no significant differences between first- or further line treatment (p=0.46) § VF occurred in 56/294 (19%) cases subsequent to LLV (Figure 1) § Median viral load at failure: 472 copies/ml (range 203-116590 copies/ml) § Mean LLV episode: 77.4 weeks (σ=68.0) § VF rate increased in TE patients (19.4%) versus TN patients (10.6%) (Figure 1)

Figure 33:: Risk of virological failure after low level viremia according to the drug. * Drug approval ≤ 2004; Virological failure rates present the percentages of the drug related low level viremia prevalence.

SUMMARY & CONCLUSION § low prevalence of LLV in patients on suppressive ART (4.8%) § VF subsequent to LLV observed in 19% of the cases § Strongest predictor for VF subsequent to LLV was a treatment regimen containing drugs approved before 2005 Figure 1: Frequency of low level viremia and subsequent Figure 4: Risk of LLV and virological failure in the AREVIR/RESINA cohort. VF according to drug § Episodes of LLV in patients treated with drugs with high potency approval and a high barrier to resistance are not predictive to VF

LITERATURE 1. Laprise, C., et al., Virologic failure following persistent low-level viremia in a cohort of HIV-positive patients: results from 12 years of observation. Clin Infect Dis, 2013. 57(10): p. 1489-96. 2. Navarro, J., et al., Impact of low-level viraemia on virological failure in HIV-1-infected patients with stable antiretroviral treatment. Antivir Ther, 2016. 21(4): p. 345-52.

Contact: [email protected] Drug Resistance Mutations (DRM) among Pregnant HIV-Positive Women in the Duesseldorf University Hospital, Germany, 2009-2016 U.E.H. Haars1, N. Lübke2, B.E.O. Jensen1, D. Häussinger1

1 Heinrich-Heine-University, Department of Gastroenterology, Hepatology and Infectious Diseases, Duesseldorf , Germany 2 Heinrich-Heine-University, Institute for Virology , Duesseldorf , Germany

Background: Combination antiretroviral Therapy (cART) has resulted in significant reduction of mother-to-child-transmission (MTCT) from 40% to 1- 2% in the last two decades. Choosing an individualized cART is one key factor for successful suppression of viral load until delivery. Thus, drug resistance testing during pregnancy before cART initiation or in case of increasing viral load is recommended. The Prevalence of DRM in pregnant women in Germany hasn't been characterised yet.

Materials and Methods: Results: From 01/2009 to 03/2016 HIV-Drug-Resistance was observed in HIV- Data of 85 HIV-positive pregnant women and 103 live births were positive pregnant Women in our special consultation for pregnancies in analysed. the Department of Gastroenterology, Hepatology and Infectious The majority ( 88%) of these women were migrants, 75% (64/85) from Diseases in the Duesseldorf University Outpatient Clinic. Subsahara Africa, 12% (10/85) South-East-Europe, 12% (10/85) from The Genotypic Resistance Testing was done concerning German- Germany and 1% (1/85) from Asia. In 34% (29/85) they had their HIV Austrian pregnancy guidelines either during their first visit when they diagnosis in their first pregnancy, in 66% (56/85) the diagnosis was were treatment naïve or being already on cART with detectable HI- upraised independently from their first pregnancy. Viral Load. In 64/85 cases (75%) resistance testing was requested, with 61/64 ( 95%) Resistance Testing was performed by using Sanger Sequencing and successful analyses. Next generation Sequencing (NGS) by means of Illumina MiSeq- The majority of the patients were infected with non-B-Subtypes (54/61, technology. Resistance Interpretation was performed by the HIV-Grade 88%), mainly 02_AG (23/61, 38%), followed by C (8/61, 13%) and A (7/61, HIV-1-Tool (www.hiv-grade.de) 11%) (Figure 2) . In 14/61 (23%) resistance tests DRM were found (Table 1), in 9/14 due to Namibia Eritrea ART-history. Patients No. 1 and 4 received PMTCT in Africa, Patient 5 was Gambia Morocco 1% 1% Thailand Ivory Coast 1% 1% Tansania 1% perinatally infected. 5/14 patients (No. 2,7,8,12 and 13) were Therapy- 1% 1% naïve with presumably transmitted DRM (tDRMr) o in Patient No. 8 DRM Angola 3% Guinea due to immunological mechanisms like APOBEC3G/F ( M184I, M230I) [1]. 3% Nigeria 5/14 patients contained a 2-class-resistance against NRTI/NNRTI. 16% Mosambik Most common mutations were: M184VI (5/14), T215Y/F/N (4/14), Y181C 3% (3/14) and K103N (3/14). Kenia NGS-analysis showed additional mutations in 2/14 patients in comparison 6% South-East Europe to Sanger: in Patient No. 1 (T215FY) and in Patient No. 8 (M230I) to Sanger 14% Kongo DRM. 6% In 2/14 Therapy-naïve patients tDRM could be shown only in NGS- sequencing : the revertant T215N in patient No. 12 and the K65R in Togo Germany 12% patient No. 2. 14% No case of MTCT has been observed. Ghana 13% Figure 1: Origin of n=85 HIV-positive pregnant women

neg. 3% n=85

n.d. 02_AG 27% 28%

K 1% 06_CPX J 2% 1% G A 6% 9% F C B 3% D 9% 1% 10% Figure 2: Table 1: Resistance Testing in n=85 HIV-positive pregnant women, Subtype distribution, n.d.= not VL=HI-Viral Load, HIV-1 Subtypes, Resistance Testing by Sanger Sequencing and Next generation done, neg.= no resistance result obtained Sequencing (NGS), PR= Protease Mutations, RT= Reverse Transkriptase mutations, ART History

Conclusions: In 23% (14/61) of all HIV-positive pregnant women in our study DRM have been observed, in 8% tDRM (5/61). The prevalence of tDRM in pregnant women in our population is lower than in general German population of HIV-positive individuals [2]. Using resistance testing by NGS resulted in the identification of additional relevant DRM compared to Sanger. Considering the importance of viral load suppression in Pregnancy and the limited amount of time to achieve this goal, the choice of cART should be optimal and take these mutations into account. Especially women from Subshara Africa harbour the risk of tDRM because of the cART regimen in High prevalence countries. The number of drug Resisance testing in developing countries is increasing. Genotypic Resistance Testing should be therefore considered for all pregnant women to optimize the success of cART and hence prevent mother to child transmission.

References: 1. Noguera-Julian M, Cozzi-Lepri A., di Giallonardo F et al. ; CROI 2014; Poster 600 2. Oette M. et al.; Intervirology 2012;55(2):154-9

Prevalence of HIV type 1 drug resistance mutations in treatment-naïve patients participating in the GARDEL Study

BACKGROUND MATERIALS AND METHODS

RESULTS

CONCLUSIONS

Author correspondence: María Inés Figueroa [email protected] CHARACTERISTICS OF SAMPLES WITH EVIDENCE OF TRANSMITTED DRUG RESISTANCE The highly effective antiretroviral therapy has changed the natural history of HIV /aids, delaying the disease progression and improving the quality of life of the infected individuals. In treated HIV-1 Type of infection Year Region of HIV-1 subtype Sex Route of TDR mutations population in Cuba, several factors might have contributed to high drug resistance levels such as residence transmission NRTI NNRTI PI prescription of suboptimal regimens containing non-boosted PI, prolonged exposure to failing therapies Chronic 26 HO B F HT K103N due to limited access to laboratory monitoring and limited options for antiviral drug substitutions if Recent 20 PR CRF24_BG F HT G190A N83D required. This might also result in the subsequent spread of drug resistant strains. The performed Recent 51 LH CRF19 M MSM M41L, M184V,T 215CY K103N studies in untreated population have shown high levels of HIV resistance to the antiretroviral therapy Chronic 30 LH CRF20_BG F HT K103N ranging from 12% to 21%. The aim of this study is determine the levels of primary HIV drug resistance in Recent 31 LH Recombinant M HT K103N newly diagnosed patients Cubans on a representative sample of the country. Recent 23 IJ Recombinant M HT L23I Chronic 56 CM CRF18 F HT K103N Recent 28 GT B M MSM D67N, M184V K103N Recent 39 PR CRF23_BG M MSM D67N, M184V, K219N Y181C

Recent 58 LH CRF19_cpx M MSM K103N STUDY DESIGN CROSS-SECTIONAL STUDY. 263 INDIVIDUALS NEWLY DIAGNOSED WITH APRIL 2013-APRIL 2014 HIV-1 INFECTION. REPRESENTATIVE FOR Chronic 45 LH CRF18_cpx M MSM K103N THE ALL COUNTRY Recent 23 LH Recombinant M MSM K101E Recent 23 LH CRF19_cpx M MSM K219N Y181C Recent 25 LH B M MSM L74V, M184V K103N Recent 24 CM CRF24_BG M MSM L210W, T215Y K103N, Y181C Recent 20 LH CRF19_cpx M MSM I47V SAMPLE PROCESSING PCR Recent 31 AR CRF20_BG F HT M184V Y181C RNA RT-PCR AND PURIFICATION 1 ML OF PLASMA WAS Recent 22 CM CRF18_cpx M MSM K101E, G190A EXTRACTION NESTED PCR PRODUCT ULTRA-CENTRIFUGED AT Recent 37 CA G M MSM Y181C 20,000 XG FOR 1 H Recent 19 CA CRF19_cpx M MSM M41L, L74V, M184V, T215SY K103N D30N

Recent 57 PR CRF19_cpx M MSM L74V, M184V, T215Y K103N D30N, N88D

Chronic 17 SC CRF20_BG F HT M184V K101E, K103N, G190A THE SEQUENCES PRODUCTS: Recent 29 CM B M MSM K103N SEQUENCE REACTION CEQ 8800 GENETIC ANALYSIS Recent 37 LH CRF18_cpx M MSM L90M SYSTEM Recent 19 HO B F HT Y181C Recent 44 HO Recombinant M MSM K219Q Y181C Recent 22 LH CRF19_cpx M HT Y181C 1–99 aa OF PR AND 1–335 Recent 49 CM CRF20_BG M MSM G190A aa OF RT Recent 39 LT G F HT K103N THE ELECTROPHEROGRAMS WERE Chronic 42 HO CRF19_cpx M MSM F77L DISPLAYED,AND ASSEMBLED. SEQUENCES Recent 40 LH B M MSM K219Q WERE MANUALLY EDITED USING Chronic 59 LH CRF19_cpx M MSM Y181C SEQUENCHER VERSION 4.10.1 AND HIV-1 Recent 23 AR Recombinant M MSM F53L SUBTYPE B STRAIN HXB2 AS A REFERENCE. From the 33 patients with TDR, 22 (66.6%) were HSH, 26 (78.8%) were diagnosed with a recent HIV-1 infection, 13 (39.4%) are from Havana and 9 (27.2%) were infected with CRF19_cpx.

HIV-1 SUBTYPING: URF THE PREVALENCE OF GENOTYPIC DRUG REGA SUBTYPING Characteristics of the new diagnostic HIV- RESISTANCE WAS ANALYZED USING THE 11% Subtype B TOOL VERSION 3 24% Subtype C 1 infected study population, April 2013- CALIBRATED POPULATION RESISTANCE (CPR) % NNRTI 2% April 2014 TOOL VERSION 6.0 AND BASED ON THE 50 SURVEILLANCE DRUG RESISTANCE MUTATION 45.4 CRF20-23-24_BG Experiments were successful for 189 (SDRM) LIST 2009 (BENNETT ET AL., 2009). 45 28% samples from 263 patients The mean age 40 CRF18_cpx at sampling was 33.5 years (17-74), the 35 NRTI 10% 30.3 80.9% of the patients were men and the 30 CRF19_cpx Subtype G 20% 4% major transmission route was the MSM 25 24.2 (80.3%).The 27.5% of patients had chronic Subtype H 20 1% infection and 72.4% recent infection. The % 50.0 15 highest number of analyzed samples was 12.1 12.1 IP 9.1 Conversely to that reported so far in the Cuban epidemic, where 45.0 10 from Havana with 38.6%, followed by the 6 subtype B was the genetic form predominating, the BG recombinants 5 eastern region of Cuba (29.6%), the 40.0 resulted the most frequent subtype detected (28%), followed by Midwest (16.9%) and finally the western 0 35.0 subtype B(24%), CRF19_cpx (20%), the unique recombinant forms M184V T215Y K219N/Q K103N G190A Y181C D30N region of the country (14.8%). The median (URF) (11%) and CRF18_cpx (10%), although other subtypes were also value viral load at the time of sampling was 30.0 The most common mutations associated with resistance to present. 58 000 RNA copies/mL (16 700-127 000) 25.0 and median CD4 count value was 371 45.5 NRTI were M184V(24.2%) followed by thymidine analogue cells/mm3 (270-573). 20.0 mutations (TAMs) such as L215Y(12.1%), K219N/Q (9%), 120 % 15.0 D67N (6%), M41L (6%). For NNRTI, K103N(45.4%), Y181C In the 17.4% (33/189) of the studied 27.3 (30.3%) and G190A (9.1%) were the most frequent 100 10.0 80 viruses, transmitted resistance mutations 17.4 mutations. The most prevalent PI was D30N (6%). 12.1 60 were detected. Simple non-nucleoside 5.0 9.1 6.1 The detection of a mutation transmitted resistance to ARVs 40 mutants contributed the highest amount 0.0 was associated with VL over 100 000 copies/mL (p = 0.025; 20 (45.5%), followed by double class Any mutations NRTI NNRTI OR = 2.464 (1.148 - 5.288)). 0 PI NRTI+NNRTI NRTI+NNRTI+PI ANY NRTI+NNRTI+ resistance against NRTI and NNRTI NRTI NNRTI PI NRTI+NNRTI The DRMs, simple mutant, or triple, were not associated with MUTATIONS PI (27.3%) and single mutants to the PRI (12. any other of the variables (type of infection, sex, sexual RI 78.8 3 30.3 12.1 24.2 9.1 1%). behavior, subtype, region or CD4 count) . CI 21.2 3 15.1 0 3 0 This study confirms the high levels of resistance in untreated population, it demonstrates the commitment of first-line therapies used in the country and could put at risk future therapies to keep or increase these figures. It highlights the need for studies to elucidate the factors that are influencing detected high levels of resistance in newly diagnosed population in order to take action or to correct the behavior or factors involved in the phenomenon. It also shows the need for resistance testing in patients who are starting the therapy. Viroseq protocol optimized for the detection of HIV-1 drug mutations in patients with low viral load. Fátima Monteiro1, Gilberto Tavares1, Marina Ferreira1, Ana Amorim1, Pedro Bastos1, Carolina Rocha1, Dina Hortelão1, Claudia Vaz1, Rosário Serrão2, António Sarmento2, Fernando Araújo1, M. Carmo Koch1 1 Molecular Biology Center, Blood Bank and Transfusion Department, Hospital S. João, Porto, Portugal 2 Infectious Diseases Department, Hospital S. João, Porto, Portugal

Background Material and Methods

Genotypic resistance testing is paramount for the Blood samples from 36 patients on HAART with a viral load monitorization of the emergence of antiretroviral drug between 20 cp/mL and 1000 cp/mL (range 36 -934 cp/mL;

resistant virus. The Viroseq HIV--11 genotyping system v v2.02.0 mean = 357 cp/mL) were collected in K3EDTA and the is an IVD assay forsequencing of HIV--11 from plasma but plasma separated 6 h after sampling and stored at -80°°C.C. only feasible if the viral load is at least 1000 cp/mL . HIV-1-1 was concentrated by centrifugation of 1 mL of plasma However, some patients have a persistent low HIV--11 at 24,000g for 1h at 4ºC. After removal of the supernatant, 1 viraemia inferior to 1000 cp/mL, being resistance testing mL of plasma was addedand the sample thoroughly and antiretroviral therapy hampered by this. So, for their homogenized. RNA extraction was performed in the clinical management, resistance testing solutions must be QIASymphonySP equipment from QIAGEN (Hilden, Germany) made available 1. With this regard we developed an in using the QIAsymphony Virus/Pathogen Mini Kit and an in house assay, adapting the Viroseq vv2.02.0 with a nested-PCR house protocol, rendering a final volume of 30 μL. The protocol. Viroseq protocol was performed according to the manufacturer instructions, followed by a nested -PCR protocol based on the previously described by NN.. Mackie et 2 Results al. The 50 μL PCR mix contained 00,5,5 μM of each primer, 11xx + Incomplete NH4 Reaction Buffer (DFS-Taq DNA Polymerase – Bioron Life Science), 00,2,2 mM of deoxyribonucleotide, 22,5,5 Sequencingand drug resistance testing was successful in Units of DFS-Taq DNA Polymeraseand 5 μL from the 70% ( (9/9/13) of the samples with a viral load 36 -200 cp/mL ; productsof the first PCR. The PCR was performed on a in 93%%((13/14) of the samples comprising 200-500 cp/mL Perkin Elmer PE9700 thermocycler and consisted on an and in 100% ( (9/9)9/9) of the samples with 500-1000 cp/mL. initial denaturationfor 5 min at 95°C, followed by 40 cycles of

Viral Load Successful Sequencing 95°C°C for30 ss;; 55°°CC for 30s, 72ºC for 120 s and a extension at 36-200 cp/mL 70% 72ºC for 7 min. PCR products were sequenced on the 200-500 cp/mL 93% 500-1000 cp/mL 100% 3130xl DNA Analyzer (Applied Biosystems)Biosystems) and analyzed in Viroseq v2.8.v2.8.

Nested PCR

Viral RNA RT-PCR PCR Sequence isolation AMPLIFICATION YES Sequencing ? analysis

Figure 1: Detection of amplification product after PCR. No detectable amplification for Sample 8 and low amplification for Sample 18.

Conclusion

Genotypicresistance testing is essential for the monitorization ofthe emergence of antiretroviral drug

Figure 2: Amplification products of samples 8 and 18 after execution of the nested PCR protocol. resistant virus being necessary the development of assays for patients with low viral loads.

1. Ryscavage P , Kelly S , Li Z , Harrigan PR, Taiwo B. Significance and clinical management of persistent low-level viremia and very-low-level viremia in HIV-1--1-infected patients. Antimicrob HIV Drug Therapy Glasgow 2016 Agents Chemother. 2014 Jul;58(7):3585-98. 23-26 October 2016 2. Mackie NE, Phillips AN, Kaye S, Booth C, Geretti AM. Antiretroviral drug resistance in HIV-1--1-infected patients with low-level viremia. J Infect Dis. 2010 May 1;201(9):1303-7. P371 The role of Presepsin (sCD14-ST) as an indirect marker of microbial translocation and immune activation P. Columpsi (1), V. Zuccaro (1), P. Sacchi (1), S. Cima (1), S. Toppino (1), S. Paolucci (2), G. Comolli (2), F. Baldanti (2), M. Mariconti (1), R. Bruno (1) 1) Dipartimento di Malattie Infettive, Fondazione IRCCS Policlinico San Matteo, Pavia . 2) Unità di Virologia Molecolare, S.C. di Microbiologia e Virologia, Fondazione IRCCS Policlinico San Matteo, Pavia

• Presepsin is a newly discovered soluble fragment of CD14 studied as a sepsis biomarker. • The mechanism of its secretion is involved in the TLR4 activation cascade and it is related to mCD14 and sCD14, which are monocyte activation markers, indirectly representing the presence of bacterial translocation. Therefore Presepsin could be employed as an immune activation marker, and it could allow for the estimation of bacterial translocation rates(1). • The aim of this study was to assess the correlations between Presepsin serum concentration and bacterial translocation, immune activation and fibrosis markers in subjects with HIV and HCV mono-infections and in HIV/HCV co-infection, compared to healthy controls.

• This is a cross-sectional study included 80 subjects followed up at the Department of infectious Diseases of Policlinico San Matteo, Pavia University. • The study population included patients with HIV mono-infection (n = 20), HCV mono-infection (n = 20), HIV/HCV co-infection (n = 20), and healthy controls (n =20). Peripheral blood was analyzed to determine the levels of Presepsin, Forkhead box 3 (Foxp3+) T cells, TGF- β1, CD14 (soluble and surface isoforms), IL-17 and bacterial translocation products. • These measurements were correlated to the severity of liver fibrosis, measured with the FIB-4 score and transient elastography.

• Presepsin concentration was significantly higher in the HIV patients (HIV monoinfected and HCV / HIV co-infected). The same group showed increased levels of sCD14 and mCD14, expression of immune activation. • Statistical analysis show a significant correlation between presepsin and both forms of CD14 only in HIV / HCV group, where the percentage of bacterial translocation and chronic inflammation is high, as shown by the significant increase in bacterial DNA levels, sCD14, mCD14 and IL-17. Presepsin is associated to FIB4 values in the HCV group.

Presepsin is a biomarker of chronic immune activation, as demonstrated by its correlations with sCD14, mCD14 and CD4+CD25+Foxp3+ lymphocytes, particularly in HIV infection. Its concentration is correlated to liver fibrosis markers, such as FIB4, particularly in HCV mono-infected patients. Considering presepsin and a direct correlation between the levels of fibrosis and an inverse correlation with Treg cells in this group, the low levels of Treg cells may be involved in increasing the state fibrosis in chronic HCV patients.

Reference 1. Yaegashi, Y., Shirakawa, K., et al.Evaluation of a newly identified soluble CD14 subtype as a marker for sepsis. Journal of Infection and Chemotherapy, 11(5), 234–238. doi:10.1007/s10156-005-0400-4 (2005). MSM: Men sex with men; HTX: heterosexual transmission. the qualitaAve variables as n (%). as median (IQR) and expressed are variables quanAtaAve The Table of the 49 cases 1. CharacterisOcs responded. responded. ﬁrst-‐line ART regimens paAents with G190A with treated paAents the in **All lower was mutaAon (4.6 vs 5.1, p=0.02). VL **Baseline *21 cases Lymphocyte CD4 nadir CD4 Lymphocyte (cel/μL) Seroconversion Ome (months)* Origin background Studies transmission HIV Median age (years) AnOretroviral therapy*** Death AIDS cases Baseline (cel/μL) lymphocyte CD4 Baseline VL (log copies/mL)** Conclusions Results Fig.1. GeneAc organizaAon of CRF19_cpx variant reference strain. Background excellent response with few cases of AIDS and excellent to ART. 4. paOents of 3. Half . cluster 2. All cases but one are related to a local Spain southern 1. CRF19_cpx variant in southern Spain, clustering in men having sex with (MSM). to AIDS progression rapid and diagnosis higher to viral load (VL) at associated pathogenic typically are infecAons recombinant from Cuba these a highly Furthermore, [1]. as described been has CRF19_cpx HIV Unlike previous studies, the variant from Malaga seems variant the studies, previous Unlike CRF19_cpx variant has emerged aﬀecAng has variant CRF19_cpx P-‐372 CharacterisOcs France ArgenOna Spain University Undergraduate school No studies/primary HTX MSM Prevalence: 2.1% cases 49 . showed the G190A resistance mutaOon.

González-‐Domenech, CM 24.8 (10.8-‐21.0) (10.8-‐21.0) 24.8 (26.3-‐41.5) 35.0

CRF19_cpx variantemergenceinaclusternaïvepatientsof 3 Hospital Carlos Haya, InfecOous Diseases, Malaga, Spain; 388 (259-‐470) 361 (254-‐416) 4.9 (4.5-‐5.4) (4.5-‐5.4) 4.9 1 southern Spain.Clinical andphylogeneticcharacterization InsOtuto de InvesOgación InsOtuto n (%)

. Here, we describe the emergence the we describe of this Here, [2]. 46 (94.0) 46 (94.0) 13 (26.5) 21 (42.9) 48 (98.0) 46 (93.9) 46 (93.9) 6 (12.2) 6 (12.2) 1 (2.04) 1 (2.04) 1 (2.0) 1 (2.0) 2 (4.0) 1 (2.0) 3 (6.1) 3 (6.1) KF716479 D KF716479 B) A) MSM naïve paOents naïve MSM Biomédica de Málaga (IBIMA), Spain; 1 ; Viciana, I less pathogenic less 1,2 ; Mayorga, M from 4 Bulgaria Hospital Costa del Sol, InfecOous Diseases, Marbella, Spain; , M Material and Methods ) were confirmed by phylogeneAc confirmed were 1 ) variant (Fig. analysis with other 195 reference sequences retrieved from LANL. a CRF19_cpx as consigned Sequences • • • • • • the analysis of HIV-‐1 genotypic drug resistance in Malaga (Spain). • used. phylogeneAc reconstrucAon inferred by maximum likelihood method (RAxML). maximum by phylogeneAc reconstrucAon inferred AddiAonally, we collected epidemiological The reliability of the clades was supported on bootstrapping four hospitals in 2011-‐2016. Genotypic test performed in 2298 naive paAents from was 3 The study was undertaken at the Virgen de la Victoria la Victoria Hospital, a reference center for de the Virgen at undertaken was study The For analysis of RT and protease resistance mutaAons and protease RT of analysis For 1,2 The subtype for each FASTA sequence provided was assigned through sequence FASTA each for subtype The Protease and reverse transcriptase (RT) genes were aligned by (RT) transcriptase reverse and Protease ; Palacios, R ; Palacios, [2] in Cuba. JAIDS.2005; 40: 532-‐537. intersubtype recombinant form (CRF19_cpx) of IdenAficaAon al. et M, Sierra a novel HIV-‐1 circulaAng ADG MM, Thomson G, Casado [1] Strongly Associated With Rapid Progression to AIDS in Cuba.EBioMedicine.2015; 2:244-‐54. ; Clavijo, E ; Clavijo, 2

KJ677057D19 Hospital Virgen de la Victoria, UGC InfecOous Diseases and Microbiology, Malaga, Spain; References Cuba Kouri V, Khouri R, Aleman Y, et al.CRF19_cpx is an EvoluAonary ﬁt HIV-‐1 Variant al.CRF19_cpx et Aleman Y, R, Khouri V, Kouri Malaga, Spain.

1,2 2 ; *Santos, J ; de la Torre, J de la ; Israel relaAons with sequences from Genbank database. ML. Only support values ≥70% are shown; values support Only ML. 49 paAents obtained by our of clustering the containing subtree and A). 2. Fig 1,2

Number of CRF19_cpx cases Fig.3. 10 12 14 16 18 PhylogeneAc tree with CRF19_cpx reference sequences, CRF19_cpx with tree PhylogeneAc 0 2 4 6 8 4 0121 0321 052016 2015 2014 2013 2012 2011 ; Jarilla, F ; Jarilla, 95 Cases of CRF 19_cpx over Ome variants 138658 , clinical 5 ; Castaño, M 93 5 Hospital de Hospital Antequera, Internal Medicine, and inmunovirological Standford algorithm v7.1.1 algorithm Standford Year *email: [email protected] 1,000 replicaOons , with 1,000 70 3 ; del Arco, A 73 B). 95 74 78 98 79 Closest phylogeneAc Closest ClustalX 146808 146882 138970 146169 138816 138436 146798 138893 138810 146950 138808 146569 146569 146764 129975 138145 138111 146937 146854 138658 146178 146741 146671 129159 146231 138445 129213 129992 129140 129979 138653 146126 146077 138650 146117 146116 146125 146912 146257 129763 129733 146964 138256 138423 138794 146842 138528 138507 138494 data. 5 ; Márquez, REGA v3.0. REGA E C A G D F B and the . G190A resistance mutation was 24 patients (48.9%)

HIV Drug Therapy Glasgow 2016

One-Step Real-Time PCR for HIV-2 group A and B RNA plasma viral load in LightCycler 2.0 Pedro Bastos1, Fátima Monteiro1, Gilberto Tavares1, Marina Ferreira1, Ana Amorim1, Carolina Rocha1, Dina Hortelão1, Claudia Vaz1, Rosário Serrão2, António Sarmento2, Fernando Araújo1, M. Carmo Koch1 1 Molecular Biology Center, Blood Bank and Transfusion Department, Hospital S. João, Porto, Portugal. 2 Infectious Diseases Department , Hospital S. João, Porto, Portugal

Background Results

Although with a lower prevalence than HIV-1, HIV-2 is The standard curve generated by the LightCycler software responsible for localized epidemies, being Portugal the non (version 4.05) presented an efficiency of 2.079 (103%), an error African country with the greatest expression of the infection. of 0.0657 and a r2 of 1.0 (Fig. 1). Besides the detection of B Clinical management of the infection is hampered by the lack of subtypes, this RT-qPCR provides a linear range between 5.03 x validated commercial RNA viral load assays, thus their in house 106 cp/mL and 5.03 x 102 cp/mL, adequate to the low HIV-2 viral development using the available equipment is mandatory. loads in plasma.

Material and Methods

Samples HIV-2 was confirmed by Innolia™ (Innogenetics, Gent, Belgium).

Blood samples were collected in K3EDTA and the plasma

separated 6 h after sampling and stored at -80°C. The BIOQ HIV- Figure 1. Standard curve generated in the LightCycler software. 2 RNA group A quantification panel (Biocentric, Bandol, France) was used as an external standard. To evaluate repeatability and reproducibility, clinical samples tested with the previous method and serial dilutions (105-102 HIV-2 RNA isolation cp/mL) of the NIBSC HIV-2 NIH-Z strain (Advanced RNA extraction was performed from 1000 μL of plasma in the Biotechnology Incorporated, Columbia, Maryland) were tested QIASymphonySP (QIAGEN, Hilden, Germany) using in replicates (Fig. 2) in the same and in independent runs with the QIAsymphony Virus/Pathogen Mini Kit and a generic different operators, with a Ct CV lower than 0.28 and SD lower protocol, rendering a final volume of 60 μL. RNA from the than 0.8 (data not shown). samples and standards was isolated under the same conditions.

HIV-2 RNA quantification (RT-qPCR) The protocol was based on the previously described by Avettand-Fenoel et al.1 Primers and probes are described on table 1. The one step RT-qPCR was performed on the LightCycler 2.0 (Roche Diagnostics, Mannheim, Germany) with Figure 2. Repeatability evaluation with NIBSC HIV-2 NIH-Z strain. the Lightcycler RNA Virus Master kit from Roche (Roche Life Sciences, Mannheim, Germany) was used. The 20 μL reaction mixture contained 0,5 μM of each primer, 0,25 μM of each Conclusion probe, 0,4 μl of Enzyme Blend and 7,5 μL of the isolated RNA. RT-qPCR cycling conditions consisted on 10 min at 60°C and 60s This assay allows us to evaluate HIV-2 A and B subtypes viral at 95°C, followed by 50 cycles of 95°C for 5s; 60°C for 50s and load in plasma with satisfactory sensibility and reproducibility, 72ºC for 10 min. supporting the clinical management of the infection. LTR region Sequence (5’-3’) Primer LTR F 5'-TCTTTAAGCAAGCAAGCGT GG-3 Primer LTR R 5'-AGCAGGTAGAGCCTGGGTGTT-3 Probe LTR P 5'FAM-CTTGGCCGGYRCTGGGCAGA-BHQ1-3 References GAG region Sequence (5’-3’) Primer gag F3 F3 5'-GCGCGAGAAACTCCGTCTTG-3 1. Avettand-Fenoel V, Damond F, Gueudin M, Matheron S, Mélard A, Collin G, Primer gag R1 R1 5'-TTCGCTGCCCACACAATATGTT-3 Descamps D, Chaix ML, Rouzioux C, Plantier JC. New sensitive one-step real-time Probe S65GAG2 5'FAM-TAGGTTACGGCCCGGCGGAAAGA-BHQ1-3 duplex PCR method for group A and B HIV-2 RNA load.J Clin Microbiol. 2014 Table 1. Primers and Probes sequences. Aug;52(8):3017-22.