CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

as baseline, treatment-enriched (those increasing in proportion to wild type from baseline to relapse) and treatment-emergent (those which were not present at baseline). Correlation analysis was performed between RAVs and sustained virologic response (SVR). Results: For treatment naïve patients, pre-treatment RAVs did not predict SVR in patients who received at least 6 weeks of DAA therapy (SYNERGY A, B, C, E, F). Similarly, no association was observed between baseline RAVs and SVR among treatment-experienced patients and those with advanced liver fibrosis (SYNERGY F) treated for 6 weeks (P=0.69). For patients receiving ultra short duration therapy (4 weeks), regardless of liver fibrosis, high level RAVs (conferring > 20 fold resistance) present prior to treatment were associated with relapse (SYNERGY G/H, P=0.03). However, when these relapsed patients were retreated with SOF and ledipasvir (LDV) for 12 weeks (85% of those previously treated with LDV/SOF had high level NS5A RAVs prior to retreatment), 91% achieved SVR (SYNERGY D). Conclusions: HCV NS5A RAVs do not predict relapse when using sofosbuvir-based treatment for at least 6 weeks or during re-treatment with standard 12 week regimens of LDV/ SOF. RAVs do appear to impact treatment response when treating with ultra-short, 4 week duration therapies.

581

Efficacy and Safety of Sofosbuvir-Based Regimens in Clinical Practice Adeel A. Butt 1 ; PengYan 2 ; Obaid Shaikh 3 ; RaymondT. Chung 4 ; Kenneth E. Sherman 5 ; for the ERCHIVES Study Group 1 Hamad Hlthcare Quality Inst, Doha, Qatar; 2 VA Pittsburgh Hlthcare System, Pittsburgh, PA, USA; 3 Univ of Pittsburgh, Pittsburgh, PA, USA; 4 Harvard Med Sch, Boston, MA, USA; 5 Univ of Cincinnati, Cincinnati, OH, USA Background: With recent approvals of multiple directing acting antiviral agents for treatment of HCV, it is important to validate and compare their efficacy in real-world settings. We compared the sustained virologic response rates (SVR), and hematologic toxicity among sofosbuvir vs. boceprevir treated persons. Methods: We used ERCHIVES (Electronically Retrieved Cohort of HCV infected Veterans) to identify HCV infected persons initiated on sofosbuvir- or boceprevir-based regimens. We excluded persons with HIV, positive hepatitis B surface antigen, hepatocellular carcinoma and missing HCV RNA. Results: Among 2,282 sofosbuvir and 1,688 boceprevir treated persons meeting the inclusion criteria, sofosbuvir-treated persons were older, more likely to be White, had higher body mass index, were more likely to have cirrhosis, diabetes, chronic kidney disease and higher HCV RNA at baseline. Overall SVR rates were 89% for the sofosbuvir group vs. 71% for the boceprevir group (P<0.001). Sofosbuvir-treated persons had higher SVR rates regardless of prior treatment status or presence of cirrhosis (P<0.001). (Table) Black race, diabetes, higher baseline HCV RNA and anemia predicted a lower SVR in boceprevir treated persons but were not associated with SVR in sofosbuvir treated persons. Presence of cirrhosis at baseline was associated with a lower SVR in both groups. Statin use was associated with a higher SVR in both groups, though the effect size was twice in the sofosbuvir group. On treatment grade 3/4 anemia (1.8% vs. 6.1%), thrombocytopenia (6.7% vs. 13.3%) and neutropenia (2.8% vs. 23.3%) were significantly less frequent among sofosbuvir- vs. boceprevir-treated persons (P<0.01 for all comparisons). Conclusions: Sofosbuvir-based regimens are more often being prescribed in persons traditionally considered to have clinical factors associated with lower SVR. Despite that, sofosbuvir-based regimens in clinical practice are associated with higher SVR rates and lower rates of grade3/4 hematologic adverse events compared with boceprevir-based regimens.

Poster Abstracts

231

CROI 2016