Digestive Diseases and Sciences (2018) 63:645–652
Ultra‑Deep Genomic Sequencing of HCV NS5A Resistance‑Associated
Substitutions in HCV/HIV Coinfected Patients
Enass A. Abdel‑hameed
· Susan D. Rouster
· Ceejay L. Boyce
· Xiang Zhang
· Jacek Biesiada
· Kenneth E. Sherman
Received: 8 September 2017 / Accepted: 18 December 2017 / Published online: 12 January 2018
© Springer Science+Business Media, LLC, part of Springer Nature 2018
Background and Aims The prevalence of naturally occurring HCV-NS5A resistance-associated substitutions (RAS) to DAA
drugs might aﬀect the response to treatment in HCV/HIV coinfected subjects. There are limited data on the frequency of
HCV-NS5A naturally occurring drug-RAS at baseline in HCV/HIV coinfected patients when ultra-deep sequencing meth-
odologies are applied.
Methods HCV-NS5A-RAS were evaluated among 25 subjects in each group. Patients were matched by age, gender, and
hepatic ﬁbrosis stage category to control for selection bias.
Results Within subtype 1a, RAS were observed in 28% of HCV monoinfected and 48% of HCV/HIV coinfected subjects.
More patients in the HCV/HIV coinfected group had clinically relevant mutations to DAA directed at NS5A.
Conclusion While the clinical signiﬁcance of this observation may be limited in highly drug adherent populations, some
HCV/HIV coinfected persons may be at greater risk of viral resistance if suboptimal dosing occurs.
Keywords HCV/HIV coinfection · NS5A drug resistance-associated variants/polymorphism · Directly acting antiviral
Direct-acting antiviral agents (DAAs) for the treatment of
hepatitis C virus (HCV) infection occur in multiple classes
including NS3A/4 protease inhibitors, NS5A inhibitors, and
NS5B (polymerase) inhibitors. Agents are characterized by
varying degrees of eﬃcacy in suppressing viral replication
and in their relative barrier to resistance [1, 2].The HCV
viruses are highly replicative, producing over 12 trillion cop-
ies of virus/day and higher HCV viral loads are observed in
those with HCV/HIV coinfection versus those with HCV
There is considerable evidence that preexistent mutations
in key catalytic sites, such as the Q80 protease variant, can
inﬂuence outcome of treatment with a simeprevir/sofosbuvir
regimen as demonstrated in the COSMOS study [4, 5]. Simi-
larly, preexistence of NS5A M28, Q30, L31, or Y93 can
reduce the 12-week sustained viral response (SVR12) to
elbasvir/grazoprevir in HCV genotype 1a-infected patients.
In clinical trials, the latter observation was so signiﬁcant that
the FDA approval of elbasvir/grazoprevir included a recom-
mendation for baseline testing for NS5A HCV resistance
variants/polymorphism prior to start of therapy as it will
aﬀect treatment decisions regarding the use of ribavirin and
duration of therapy [6, 7].
Most clinical trials have evaluated resistance-associated
substitutions (RAS) using relatively less sensitive Sanger
sequencing methods or next-generation sequencing with
high cutoﬀ thresholds in the range of ≥ 10% detection .
In this study, we applied ultra-deep next-generation sequenc-
ing methodology to determine the representation of RAS
down to 1% of the population level, permitting direct com-
parison of HCV/HIV coinfected patients with those with
HCV alone. We hypothesized that HCV/HIV coinfected per-
sons might have impaired clearance of lower ﬁtness variants
which would result in greater representation of those vari-
ants in the total viral population.
* Kenneth E. Sherman
University of Cincinnati College of Medicine, 231 Albert
Sabin Way, Cincinnati, OH 45267-0595, USA
Department of Environmental Health, University
of Cincinnati, Cincinnati, OH 45267, USA