CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

Conclusion: In the SCCS, 5.3% of patients with chronic hepatitis C had a VLVL in the absence of antiviral therapy. Morbidity and mortality were comparable in patients with and without VLVL. Interestingly, and counterintuitively, many had conditions associated with immunosuppression. Missing patients at risk of disease progression may limit the applicability of HCV antigen testing. 580 EVALUATION OF DBS HCV RNA QUANTIFICATION AND GENOTYPING IN RESOURCE LIMITED SETTINGS Jayaseelan Boobalan 1 , Thongadi R. Dinesha 1 , Sathasivam Sivamalar 1 , Selvamurthi Gomathi 1 , Rajendran Revathi 1 , Aylur K. Srikrishnan 1 , Amrose Pradeep 1 , Shruti H. Mehta 2 , Sunil S. Solomon 2 , Pachamuthu Balakrishnan 1 , Shanmugam Saravanan 1 1 YR Gaitonde Center for AIDS Research and Education, Chennai, India, 2 Johns Hopkins University, Baltimore, MD, USA Background: High burden of HCV infection in people who inject drugs (PWID) in resource limited settings (RLS) like India coupled with low access to HCV services necessitates an urgent need for evaluating less invasive DBS based HCV virological assays for appropriately tailored therapy and successful treatment scale up programs in RLS. Methods: Paired 36 plasma/DBS samples were collected from PWID. Two blood spots were excised into 1.7 mL of Abbott lysis buffer. RNA extracted from plasma and DBS according to the standard 0.5 mL HCV RNA Abbott extraction protocol (Abbott Molecular Inc, IL, USA). In addition, patient with >3 log 10 IU/mL HCV RNA were further subjected to in-house one-step core/E1 RT-PCR followed by Sanger’s sequencing. All the values were log transformed and analysed on GraphPad Prism 5.0, A p <0.005 was considered significant. Results: There was a good correlation between standard Abbott plasma and DBS HCV assay (r = 0.97, p<0.001).Median PVL of standard plasma and DBS HCV RNA by Abbott Real-time PCR was 4.45(IQR 2.54-5.43) and 4.18 (IQR 2.24-5.74) log IU respectively. The mean difference between plasma and DBS HCV RNA assays were 0.250 and the upper and lower 95% Limit of Agreement -0.7321 to 1.23374. Out of 36 samples, 21 (58%) [Median = 5.74, (IQR=4.96-5.91)] of themwere able to do HCV genotyping and majority of themwere genotype 3b (42.8%) followed by 6xa (33%). Conclusion: This study supports the use of DBS as an alternate to plasma for the reliable quantification of HCV RNA and genotyping. Therefore, the DBS specimens could be effectively used in resource-limited settings for therapeutic and research purposes. 581 CLUSTERING OF HEPATITIS C VIRUS INFECTION AMONG PEOPLE WHO INJECT DRUGS IN BALTIMORE Oluwaseun Falade-Nwulia 1 , Shruti H. Mehta 1 , Mark Sulkowski 1 , Carl A. Latkin 1 , David L. Thomas 1 , Zach Downing 1 , Rachel Latanich 1 , Gregory D. Kirk 1 , Rebecca Rose 2 , Stuart C. Ray 1 , Oliver Laeyendecker 3 1 Johns Hopkins University, Baltimore, MD, USA, 2 Bioinfoexperts, LLC, Thibodaux, LA, USA, 3 NIAID, Baltimore, MD, USA Background: The availability of effective, safe, oral direct acting antivirals (DAAs) for hepatitis C virus (HCV) treatment has increased interest in treatment as prevention (TasP) for HCV infection among people who inject drugs (PWID). Identifying characteristics of individuals in high transmission networks would provide critical information for the development and implementation of effective, targeted HCV TasP strategies. Methods: The AIDS linked to the IntraVenous Experience (ALIVE) cohort has followed current and former PWID in Baltimore since 1988. Sequencing of the HCV core/E1 region was performed on HCV viremic samples from the most recent study visit attended by ALIVE participants between August, 2005 and December, 2016 (n=600). Individuals with HCV genotype 1 infection (85%) were included in the analysis. Analyses of pairwise distances among participant HCV sequences was performed using the TN93 model. Clustering of HCV infection was defined as ≥ 2 participants with HCV sequences more similar than a previously-determined genetic distance threshold of 4%. Logistic regression was used to assess sociodemographic factors associated with being in an HCV cluster. Results: Among 512 HCV genotype 1 viremic PWID, the median age of participants was 54 years, 68%male, 87% Black, and 38% HIV infected. Of the 425 genotype 1a samples evaluated, 31%were grouped in 2 large clusters (cluster 1: n=42 & cluster 2: n=66), a triad (n=3), or pairs (n=24). Among 87 PWID with HCV genotype 1b, 40%were in 2 clusters (cluster 1: n=14 & cluster 2: n=8), 3 triads (n=9), and 2 pairs (n=4). 342 (67%) of evaluated samples

clinical information including scheduled clinic visits, liver fibrosis staging by FibroScan or FibroSure, initiation and completion of treatment, RNA testing for sustained viral response (SVR) was abstracted via chart review. Results: Overall , 446 ED patients with HCV infection were identified and 384 (86%) had chronic HCV infection and were not already in care. The majority were male (66%), African American (76%) and aged ≥ 50 years (66%). Of those 384 patients eligible for the LTC program, 165 (43%) were linked to care. Of 165 linked to care, 93% (n=154) had liver fibrosis staging information available and 66 (42.9%) patients had advanced (metavir stage F3 and F4) liver fibrosis. Among 154 with fibrosis staging, 84 (55%) initiated HCV treatment and 89% (n=75) completed the treatment regimen. Among those who completed HCV treatment, 63 (84%) had HCV RNA testing result available and all (100%) achieved SVR after treatment. LTC was positively associated with older age and being black race (p<0.05). There was no difference in recipient or completion of treatment, or SVR by advanced fibrosis. Conclusion: A program to improve LTC substantially improved HCV cure rates even among patients attending EDs. However, further improvement in LTC and treatment rates are needed to achieve the ultimate success of the program and more work is needed to identify newmethods to link additional patients through increasing patients’ level of HCV

Poster Abstracts

579 VERY LOW HEPATITIS C VIRAL LOADS IN ABSENCE OF THERAPY: IMPACT ON HCV ANTIGEN TESTING Barbara Bertisch 1 , Matteo Brezzi 1 , Francesco Negro 2 , Beat Muellhaupt 3 , Patrizia Kuenzler-Heule 4 , Patrick Schmid 4 , Olivier Clerc 5 , Alberto Moriggia 6 , Darius Moradpour 7 , Olivia Keiser 1 1 University of Geneva, Geneva, Switzerland, 2 University Hospitals of Geneva, Geneva, Switzerland, 3 University Hospital Zurich, Zurich, Switzerland, 4 St. Gallen Cantonal Hospital, St. Gallen, Switzerland, 5 Cantonal Hospital Neuchatel, Neuchatel, Switzerland, 6 Epatocentro Ticino Foundation, Lugano, Switzerland, 7 Lausanne University Hospital, Lausanne, Switzerland Background: HCV antigen testing is a less expensive alternative to PCR but it has a low sensitivity for very low viral loads (VLVL ≤ 3,000 IU/ml). So far, data on VLVL came from treated persons and results are therefore not applicable to screening in a treatment-naïve population. Methods: We assessed the prevalence and analyzed predictors of VLVL by logistic regression in treatment-naïve participants in the Swiss Hepatitis C Cohort Study (SCCS). We analyzed if the last viral load after the first documented VLVL continued to be low; and we compared prevalent as well as incident cirrhosis and mortality in persons with VLVL. For persons with VLVL and cirrhosis, we extracted information on immunosuppressive conditions from the patient charts. Results: We included 2,460 persons (45.5% of 5,409 persons enrolled in the SCCS) with HCV viral loads measured in the absence of antiviral therapy. Overall, 5.3% of these ever had a VLVL. Age ≤ 40 years (aOR 1.90, 95% CI 1.26-2.85) was associated with VLVL, while gender, HCV genotype and intravenous drug use were not. In 21 of the 70 persons with viral load available after VLVL, the last viral load was still ≤ 3,000 IU or undetectable, including 8 persons with spontaneous HCV clearance during long-term follow-up. Among the 130 persons with VLVL, 24 had prevalent or incident cirrhosis. All of them had either excessive alcohol consumption, HIV coinfection, organ transplantation or other potential immunosuppressive conditions. The mortality rate was comparable in persons with and without VLVL.

CROI 2018 213