Resistance-associated substitutions and response to treatment in a chronic hepatitis C virus infected-patient: an unusual virological response case report

Hepatitis C virus (HCV) infects around 71 million people worldwide [1], and is a leading cause of chronic liver diseases including cirrhosis and hepatocellular carcinoma. Nevertheless, many chronically infected patients remain asymptomatic over the years and only become aware of their infection once they develop cirrhosis [2].

Although these DAAs seem to be very effective in curing most of the infected patients, several obstacles remain for their broad implementation including limited access to therapy, and the difficulty in treating patients with comorbidities and/or those with advanced cirrhosis or HCC [3].

Concomitantly with the development of these drugs, and taking into account the evolutionary rate of this virus, resistant variants have emerged. These variants are defined by the presence of one or more specific resistance-associated substitutions (RASs). The detection of basal RASs to DAAs in naïve patients could be important in predicting the treatment outcome in some patients exhibiting failures to DAA-based therapies (5–10%) [4, 5]. Additionally, some reports show that the presence of RASs as minority or low frequency variants in the viral population could influence the final prognosis of the patient [6].

Here, we report a case of a chronic hepatitis C infected patient followed and treated unsuccessfully through the years with no other complication beyond the infection with this virus. These results not only illustrate the difficulties that clinicians go through when treating these patients, but also suggest a possible molecular contributing factor to the treatment failure.

In this study we report the presence of RASs to HCV NS5A and NS5B inhibitors as minority variants (less than 1%) in a relapser patient, both before and after treatment with DAAs. It is of note that we did not detect any basal RASs by Sanger sequencing.

Given that the patient did not present any comorbidities, no concomitant medications or significant fibrosis before therapy started and that he presented perfect adherence to it, he was expected to respond to this IFN-free regimen (SOF/LDV + RBV) and to reach an SVR. However, the patient relapsed 12 weeks after EOT.

The presence of clinically relevant RASs has been associated to lower SVR rates when found pre-treatment (basal RASs). It is commonly accepted that they must be present as high frequency variants (> 15%) within the viral populations to affect treatment outcome in a significant way [4, 5, 10]. Still, controversy exists about whether detection of mutations below that threshold, may have an impact in treatment outcome [11]. A fact that is clear though, is that many relapser patients exhibit clinically relevant RASs (detected by Sanger sequencing) at the time of virologic failure [12‐14]. However, there are only a few reports showing that those RASs are selected from pre-existing low frequency variants which emerged as dominant mutations after DAA pressure [6, 15], highlighting the importance of monitoring RASs which go undetected through Sanger sequencing.

Importantly, the RASs found here were previously undetected by Sanger sequencing but they were already present at low frequencies. In addition, they have increased their frequency after DAA pressure (See Fig. 2). In this case, the different RASs found did not become dominant after treatment and the patient still relapsed. The fact that the post-treatment sample was taken 12 weeks after the EOT raises the question of the persistence of the RASs when the selective pressure was removed for almost 3 months. Especially if certain RASs could affect viral fitness, they might go back to low frequencies even if during treatment, they became high frequency variants. However, Paolucci et al, (2017) [4] described the presence of low frequency RASs in 60% of patients subjected to DAA therapy who achieved SVR. Their work suggests that these variants are not always subsequently selected for and might not influence therapy outcome. Nevertheless, the RASs detected in the relapser patient here could have played a role in the virologic failure given that no other risk factor was found to explain it.

Even though we did not analyze viral diversity after the second relapse to DAA-therapy (SOF/VEL), it should be taken into account that at this point the patient had already progressed to an advanced fibrosis state (F3), which led to an infection more difficult to treat. Additionally, some of the RASs detected after SOF/LDV/RBV EOT had increased their frequencies, and it is already documented that many NS5A RASs persist for long periods in the infected individual [16]. Since the RASs here detected also confer resistance to VEL, it might also be possible for them to have influenced the second DAA therapy if they were still present when the treatment started.

Our study addresses one possible contributing factor to an unusual therapeutic response with no clear association to any other clinical aspect. We cannot discard that other factors might have also contributed to the final outcome of the patient. Despite the existence of only a few reports linking single-nucleotide-polymorphisms (SNPs) near IFN-λ3 gene (formerly known as IL28B) with response to IFN-free regimens [17, 18], the patient presented in this report harbored unfavorable genotypes both in rs12979860 (TT) and rs8099917 (GG). Furthermore, the presence of highly represented amino acid substitutions (not RASs) with potential predictive value of treatment failure [19], such as Q309R in NS5B, found as majority variant both pre and post SOF/LDV treatment, might also jointly account for the observed outcome. In addition, this work seeks to stress the contribution of NGS in clinical centers, in particular for those hard-to-treat patients that have relapsed at least once to a DAA-based therapy. To this aim, we recommend storing plasma samples before and 12 weeks after each DAA treatment in case a patient relapses, in order to have samples in hand if needed for NGS studies. Nowadays, NGS is becoming a more commonly used and affordable tool that might have an impact in better therapeutic decisions and potentially improving clinical outcomes. Unfortunately, a limitation of this work is that minority variants were not taken into account for the choice of treatment. This is mainly explained by the fact that currently, the only treatment financed by Uruguayan health insurance is the use of SOF plus daclatasvir. Any other optimal DAA treatment would need to be specifically requested and its financing could be denied.

This report demonstrates that some patients might relapse after a DAA-based therapy even when RASs (pre- and post-treatment) are detected in very low frequencies (< 1%). This fact highlights the importance of using NGS approaches to better understand treatment outcomes and tailor future therapy regimens that might be better suited to some DAA-experienced patients with no other well-documented risk factors that might account for a treatment failure. Increased awareness of this association may improve detection and guide towards a personalized HCV treatment, directly improving the outcome in hard-to-treat patients.