CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

(Missing or NC=Failure). Costs included the direct cost of antiretrovirals (Spanish official prizes) plus those related with outpatient visits, hospital admissions and resistance tests (local prizes). Efficiency or cost-effectiveness was the ratio between costs and effectiveness for the base case scenario and for the most (higher 95% CI for effectiveness and lower 95% CI for cost) and less favorable (lower 95% CI for effectiveness and higher 95% CI for cost) scenarios as a sensitivity analysis. Results: 501 patients (329 ATR and 172 EPA) were included. Median age was 35 years, 92%were males, 8%were co-infected with HCV,17% had a CD4+ cell count < 200 and 23% had viral load >100,000 copies/ml (34% in the ATR group vs 1% in the EPA group; p<0.0001). Median duration of assigned regimen was 3.7 and 1.6 years for the ATR and EPA respectively due to late enrollment of EPA patients. Almost all patients (99% and 100% in the ATR and EPA arms respectively) completed at least one year of follow-up. Response rate at 48 weeks were 77% and 88% for ATP and EPA (difference of 9.2%; 95% CI 5.9% to 25.8%; p < 0.0042) respectively. Virological failure and interruptions for tolerance problems were 3% and 19% the ATR arm and 2% and 9% in the EPA arm respectively (p<0.001). In the case base scenario cost per responder at 48 weeks (cost-effectiveness or efficiency) was 12,177 euros in the ATR arm and 9,366 euros in the EPA arm (ratio of 0.85 or 15% lower). Similar trends were observed in the less and most favorable scenarios. Conclusions: The cost per responder at 48 weeks (cost-effectiveness or efficiency) of EPA in naïve patients was 15% lower than ATR mainly driven by a better tolerance. Our data supports the recommendation of EPA as opposed to ATR when baseline viral load is below 10 5 HIV RNA copies/ml of plasma. 946 Correlations of Pre-ART HIV-DNAWith Outcome in First-Line Treated ART Patients Francesca Ceccherini-Silberstein 1 ; Alessandro Cozzi-Lepri 2 ; Esther Merlini 3 ; Giulia Marchetti 4 ; Maria R. Capobianchi 5 ; Andrea De Luca 6 ; Nicola Gianotti 7 ; Andrea Antinori 5 ; Carlo Federico Perno 8 ; Antonella d’Arminio Monforte 9 ; for the ICONA Foundation 1 Univ of Rome Tor Vergata, Rome, Italy; 2 Univ Coll London, London, UK; 3 Univ of Milan, Milan, Italy; 4 Clinic of Infectious Diseases, Univ of Milan, San Paolo Hosp, Milan, Italy; 5 Inst Nazionale Malattie Infettive “L. Spallanzani”, Rome, Italy; 6 Univ of Siena, Siena, Italy; 7 San Raffaele Scientific Inst, Milan, Italy; 8 Univ of Rome Tor Vergata, Roma, Italy; 9 Infectious Diseases Clinic, San Paolo Hosp, Univ of Milan, Milan, Italy Background: Quantification of HIV-1 DNA reflects the size of viral burden in HIV-1 infected patients (pts). By using a commercial assay, we aimed to investigate the correlation of pre-ART (baseline, BL) HIV DNA with the BL viro-immuno-clinical status and on the viro-immuno-clinical response to ART in pts starting their first regimen. Methods: HIV + pts of the ICONA cohort, starting 1st-line ART, for whom PBMC or blood sample was stored at BL were analysed. Total HIV-DNA was quantified by using a modified version of the Cobas HIV-1 test (Surdo et al 2015). Results were normalized by CD4 + cells nr. Associations between BL DNA levels and pts BL characteristics were evaluated using chi-square/signed-rank test and Pearson correlation. In a subset of pts starting “modern” ART (after 2004), standard survival analysis was used to examine the association between BL HIV-DNA and pre-defined time to event endpoints: viral load (VL) ≤50 cp/mL, virological rebound defined by a confirmed VL>50 cp/mL after VL≤50 cp/mL, gain in CD4 count >200 cells/mm 3 after ART and AIDS diagnosis or serious non-AIDS or death. Kaplan-Meier curves and Cox regression models were also used. Results: We included 607 pts (23% female, 38 yrs median age, median [IQR] CD4 = 288 [144-401], who started ART on median [IQR] = 2010 [2002-2011]. BL median [IQR] HIV-DNA and HIV-RNA was 10574 [3208-38218] cp/10 6 CD4 + and 4.83 [4.31-5.33] log cp/mL, respectively. According to BL HIV-DNA levels (divided in 3 groups: 10-1000 cp/10 6 CD4 + , n=69, 1000-10000 cp/10 6 CD4 + , n=224, and >10000 cp/10 6 CD4 + , n=314), a strong significant correlation (p<0.001) was observed with BL HIV-RNA (Pearson rho=+0.41), CD4 (-0.48) and CD4/CD8 ratio (-0.40) and less strong (p<0.03) with IL-6 (+0.15), sCD14 (+0.11) and CD8 (-0.09). By wk 48, 393/607 (65%) achieved HIV-RNA ≤50 cp/mL (290/395, 73% in patients starting ART after 2004). Within this subset, median (95% CI) times to a HIV-RNA ≤50 cp/mL were 7 (4-11) mo in people with 10-1000 HIV-DNA cp/10 6 CD4 + , 8 (5-11) mo in those with 1000-10000 cp/10 6 CD4 + vs. 10 (7-15) mo in those with >10000 cp/10 6 CD4 + (p=0.0008). Unadjusted and adjusted hazard ratios of the pre-defined outcomes from fitting the Cox models are shown in Table. Conclusions: Pre-ART HIV-DNA content in CD4 + cells strongly correlated with BL viro-immunologic parameters and inflammation markers. BL HIV-DNA was also found to predict virological and clinical outcome of 1 st -line ART, although not independently of plasma VL and CD4 count. Table hazard ratios of various outcomes per log10 10 6 /CD4 HIV-­‐DNA higher levels in patients starting 1 st line ART after 2004

Hazard Ratio (95% CI) p-­‐value

Outcomes

Adjusted (a)

Adjusted (b)

Adjusted (c)

Unadjusted

Viral load ≤50 cp/mL

0.71 (0.62, 0.82)

0.73 (0.63, 0.85)

0.86 (0.72, 1.02)

0.89 (0.74, 1.06)

Poster Abstracts

P<.001

P<.001

P=0.074

P=0.197

CD4 count gain >200 cells/mm 3 above pre-­‐ART

1.01 (0.88, 1.16)

1.00 (0.86, 1.16)

0.88 (0.74, 1.05)

0.87 (0.73, 1.04)

P=0.896

P=0.994

P=0.156

P=0.133

Viral rebound >50 cp/mL

2.14 (1.41, 3.25)

1.99 (1.29, 3.09)

1.45 (0.88, 2.39)

1.23 (0.71, 2.12)

P<.001

P=0.002

P=0.142

P=0.459

AIDS, serious non-­‐ AIDS and death

2.14 (1.13, 4.05)

2.08 (1.00, 4.33)

1.49 (0.69, 3.20)

1.53 (0.59, 3.95)

P=0.019

P=0.051

P=0.309

P=0.380

(a) Adjusted for calendar year of ART initiation and type of regimen started (b) Adjusted for viral load and CD4 count at ART initiation (c) Adjusted for all factors in footnotes a-­‐b, + age, smoking, HCV status, IL-­‐6 and sCD14.

947 Comparable Viral Decay in Dual and Triple Dolutegravir-Based Antiretroviral Therapy Omar G. Sued 1 ; Maria I. Figueroa 1 ; Maria J. Rolon 1 ; Patricia Patterson 2 ; Dannae Brown 3 ; Ana M. Gun 4 ; Michael Aboud 5 ; KimberleyY. Smith 6 ; Pedro Cahn 1 1 Fundación Huésped, Buenos Aires, Argentina; 2 Hosp Juan A. Fernández, Buenos Aires, Argentina; 3 ViiV Hlthcare, Abbotsford, Australia; 4 Centro Medico Huesped, Buenos Aires, Argentina; 5 ViiV Hlthcare, Brentford, UK; 6 ViiV Hlthcare, Research Triangle Park, NC, USA Background: Drug-sparing strategies have been explored lately, aiming to improve tolerability and adherence, reduce toxicity, and costs. Two studies have shown non inferiority of dual therapy in ARV –naïve HIV-1 infected individuals (NEAT and GARDEL). The PADDLE study showed that a dual therapy regimen based on dolutegravir plus lamivudine (DTG/3TC) induced a rapid viral decay in treatment naïve patients with screening pVL <100K (EACS 2015). Our objective is to compare differences between plasma viral load (pVL) change at each time point with a dual therapy, DTG/3TC, to triple therapy regimens used in the SPRING-1 (DTG 50 mg +2NRTIs) and SINGLE study (DTG plus abacavir/lamivudine), in patients with baseline (BL) pVL < 100,000 copies/mL. Methods: In PADDLE (n=20), pVL was tested at BL, days 2, 4, 7, 10, weeks 2, 4, 8, 12, 24 and thereafter. In SINGLE (n=280) it was measured at BL, wk 2, 4, 8, 12, 16, 24 and thereafter. In SPRING-1 (n=39) pVL was measured at BL, wk 1, 2, 4, 8, 12, 16, 20, 24 and thereafter. Change in pVL vs. baseline was calculated only for time points with data for the three studies (Weeks 2, 4, 8, 12 and 24). Effects of time and treatment, was analyzed by two-way ANOVA, followed by Tukey-HSD post-hoc test.

401

CROI 2016