CROI 2017 Abstract e-Book

Abstract eBook

Oral Abstracts

gait speed. An inflammatory index score was constructed from serummeasures of interleukin-6 (IL-6) and soluble tumor necrosis factor-α receptor-1 (sTNFR1). Markov transition models were used to assess the relationship of sociodemographics, chronic comorbid disease, HIV clinical factors and inflammation to transitions between frailty states. Results: Among 1353 ALIVE participants with 9559 frailty transitions, 33%were HIV positive. Younger age, higher educational attainment, employment, and reduced chronic disease comorbidity were significantly associated with reduced frailty progression and greater frailty recovery. Adjusting for sociodemographics, substance use and comorbidity, HIV virologic suppression, elevated CD4 nadir (>500) and absence of a prior AIDS diagnosis were all significantly associated with reduced frailty progression and improved frailty recovery. Likelihood of frailty transition for these less advanced HIV disease states was similar to HIV seronegative status. Adjusting for sociodemographics, comorbidity, substance use and HIV disease stage, for each standard deviation decrease in inflammatory index score, there was a 22% decreased likelihood of frailty progression (aOR 0.78; 95% CI, 0.65, 0.92) and 29% increased likelihood of frailty recovery (aOR 1.29; 95% CI, 1.08, 1.53). Conclusion: HIV virologic suppression with early ART, improved socioeconomic environment, prevention and reduction of chronic disease comorbidity, and reduced inflammation may reduce frailty progression and promote frailty recovery; these interventions could reduce hospitalization and improve survival for persons aging with HIV. 134 INCIDENCE OF HEPATITIS C AMONG HIV-INFECTED MEN WHO HAVE SEX WITH MEN, 2000–2015 Antoine Chaillon 1 , Christy M. Anderson 1 , Thomas C. Martin 1 , Edward R. Cachay 1 , David L. Wyles 2 , Davey M. Smith 1 , Susan J. Little 1 , Richard S. Garfein 1 , Natasha Martin 1 1 Univ of California San Diego, San Diego, CA, USA, 2 Denver Health, Denver, CO, USA Background: International reports of a hepatitis C virus (HCV) epidemic among HIV-infected men who have sex with men (HIV+MSM) associated with recreational drug use and with sex are causing concern. However, little is known about the HCV epidemic among MSM in San Diego. We assess HCV incidence among HIV+MSM in San Diego in relation to injecting drug and methamphetamine use. Methods: We performed a retrospective cohort analysis of HCV incidence among HIV+MSM attending the largest HIV clinic in San Diego (UCSD Owen Clinic) from 2000-2015. Incident HCV infection was assessed among HIV+MSM with a baseline negative anti-HCV test between 2000 and 2015, and defined as any new positive anti-HCV or HCV-RNA test after the start of follow-up. Group risks were defined as individuals who reported ever injecting drug use (IDU) or using methamphetamine. Results: A total of 2,396 MSM, who were initially HCV uninfected and had at least 1 further test during a median of 5.5 years of follow-up (IQR 2.8-9.2), were included in the incidence analysis. Overall, 149 HCV seroconversions occurred over 12,560 person-years of follow-up (incidence rate = 1.19/100py, [95%CI 1.01-1.39]), which increased over time (p=0.027, Fig.1A). Individuals were tested a median 3 times (interquartile range[IQR] 2-4) with a median testing interval of 1.2 years (IQR 0.6-2.2). Incident cases were identified on average of 10.6 years (IQR: 5.7-17.5) from HIV diagnosis and 3.6 years (IQR 1.5-6.4) from the first HCV negative test. Among individuals who seroconverted for HCV, 13.4% (20/149) denied IDU and methamphetamine use. HCV incidence was significantly higher among HIV+MSM reporting IDU compared to those not reporting IDU (2.6/100py vs 0.97/100py, p<0.001), with no evidence of an increasing trend over time (Fig.1B). HCV incidence was also significantly higher among HIV+MSM reporting ever using meth compared to those denying meth use (1.53/100py vs 0.52/100py, p<0.001) with a significant increase of HCV incidence over time (p<.001, Fig.1C). Conclusion: These findings suggest that HCV incidence is increasing among HIV+MSM in San Diego. These rates are similar to London and other major European cities, and double that observed in the US Multicenter AIDS Cohort Study. This study also documented incident HCV infection among HIV+MSM who do not inject drugs and an increased HCV incidence among individuals reporting meth use. Further work is needed to explain this trend and identify prevention strategies required to control the epidemic.

Oral Abstracts

135 MODELING HIV-HCV EPIDEMIOLOGY IN THE DAA ERA: THE ROAD TO ERADICATION Victor Virlogeux 1 , Laurent Cotte 2 , Pascal Puglièse 3 , Isabelle Poizot-Martin 4 , Marc-Antoine Valantin 5 , Lise Cuzin 6 , Jacques Reynes 7 , Eric Billaud 8 , Thomas Huleux 9 , for the DatAIDS Study Group 1 INSERM, Lyon, France, 2 Hospices Civils de Lyon, Lyon, France, 3 CHU de Nice, Nice, France, 4 Assistance Publique–Hôpitaux de Marseille, Marseille, France, 5 Assistance Publique–Hôpitaux de Paris, Paris, France, 6 CHU Toulouse, Toulouse, France, 7 CHU de Montpellier, Montpellier, France, 8 CHU de Nantes, Nantes, France, 9 Cntr Hosp de Tourcoing, Tourcoing, France Background: HCV Direct-Acting Antiviral (DAA) treatment uptake has drastically increased in HIV-HCV coinfected patients in France, resulting in HCV cure in more than half of patients at the end of 2015. We used model projections to estimate the impact of DAAs on HIV-HCV epidemiology over the next decade. Methods: The model was based on epidemiological data from the French DatAIDS cohort. Eight risk groups were considered: high-risk (HR) and low-risk (LR) MSM and males/ females heterosexuals, IVDU or patients with other risk. To model HIV-HCV epidemiology we used a mathematical compartmental deterministic model calibrated using the 2012- 2015 observed incidence, prevalence and treatment coverage. Figures of the undiagnosed HIV-HCV epidemic were estimated from a previously published model (Supervie AIDS 2014). The impact of scaling-up DAA on HCV prevalence was investigated across the different risk groups. Results: On January 1st, 2016, 156,811 patients were estimated to be infected with HIV in France (under care: 131,861) of whom 7,939 (5.1%) had an active HCV infection with a detectable serum RNA (under care: 7,216 (5.5%)). Assuming a treatment coverage (TC) of 30%/year (observed rate in 2015), active HCV prevalence among patients under care dropped to 1.31% in 2021 and to 0.55% in 2026. Sub-analyses showed similar results in most risk groups, including LR MSM. Due to higher acute infection and reinfection rates, the predicted prevalence in HR MSM increased from 2.39% in 2016 to 2.46% in 2021 and 3.96% in 2026. Increasing TC in HR MSM to 50/70% per year decreased the prevalence in this group to 1.10/0.50% in 2021 and to 0.86/0.19% in 2026. With the current TC of 30%, the mean delay to reach <50 active infections per group was 10.5 years for every risk group except HR MSM. This threshold was reached at 5.5 years in HR MSM only when increasing TC to 70% in this group. At 30% TC, undiagnosed patients will account for 26.4% of patients with active HCV infection in 2021 and for 43.3% in 2026. Conclusion: Our model suggests that DAA based treatments could nearly eradicate HIV-HCV coinfection in France within 10 years in most of the risk groups, including LR MSM. Consequently acute infections and reinfections in HR MSM and undiagnosed HIV-infected patients will account for the majority of infections in the future. Eradication in these 2 groups will require increased treatment coverage of acute infections in HR MSM and increased engagement in care for undiagnosed infections.

55

CROI 2017