Impact of Recent Drug Use on the Efficacy of Elbasvir/Grazoprevir for HCV-infected People on Opioid Agonist Therapy

Juan Macías; Francisco Téllez; Ignacio de los Santos; Luis E. Morano; Dolores Merino; Rafael Granados; Francisco Vera-Mendez; Alejandro González-Serna; Esther Campo-Moneo; Lucio García-Fraile; Federico García; Luis M. Real; Juan A. Pineda


J Viral Hepat. 2021;28(6):878-886. 

In This Article


In the present study, we found that the overall SVR rates achieved with EBR/GZR were high in real-world conditions of use. All study groups, non-PWID and PWIDs, with and without OAT, reached greater than 90% SVR rates. However, PWID with recent drug use are a critical group of patients that achieve suboptimal response rates with EBR/GZR, as with other combinations.[9] The main reason for low SVR among recent drug users is voluntary drop-out. Higher relapse rates were also observed, potentially indicating incomplete adherence to therapy.

In the C-EDGE CO-STAR trial, adherence to EBR/GRZ was similar to that of the rest of the C-EDGE programme.[5–8] Very high adherence rates were reached in the C-EDGE CO-STAR trial. This was most likely due to the intensive adherence monitoring within the trial. In clinical practice, such intervention is difficult to implement. In the present study, PWID with OAT, managed in routine clinical practice, achieved SVR rates lower than those reported in the C-EDGE CO-STAR trial. Among the reasons for not achieving SVR in our study among PWID on OAT, the frequency of viral recurrence, relapses or breakthrough, was similar to the C-EDGE CO-STAR trial. However, the frequency of losses to follow-up among PWID on OAT was higher than in C-EDGE CO-STAR trial. These results are in agreement with our previous report on interferon-free DAA combinations used before 2017 in the HEPAVIR-DAA and GEHEP-MONO cohorts.[9] The herein reported results are also in agreement with other cohort studies.[13,14] Lower SVR rates among PWID on OAT were related to losses to follow-up. Hence, treatment with EBR/GZR for PWID on OAT needs to be complemented with some sort of strategy to ensure adherence.

Recent drug use was very frequent among PWID on OAT. Namely, 28% of them referred recent drug use. In the C-EDGE CO-STAR trial, more than 50% of the participants had positive results in urinary drug screening at baseline and during the follow-up.[8] However, our figure could be an underestimation of drug use in the study patients, because drug use was self-reported. However, PWID identified as recent drug users might be those with more disruptive behaviour associated with drugs leading to drop-out. This may explain the large effect of recent drug use on the response to EBR/GZR. A number of potential interventions to improve healthcare retention of PWID have been described, as co-location of OAT and DAA therapy,[14] peer support or patient navigator,[15] and cash incentives.[16] Further interventions, aside from EBR/GZR or any other DAA combination, are clearly needed to support the continuity of treatment and follow-up.

In the present study, the SVR rates for HCV genotype 1a were lower than for other genotypes. Non-optimal management of EBR/GZR in genotype 1a might account for that lower efficacy. Among genotype 1a infected patients with baseline HCV RNA ≥800,000 IU/ml, nearly half of them were not managed according to guidelines in force in Spain. However, only one out of three viral recurrences observed among patients with genotype 1a infection and high viral load could be regarded as non-optimal management. The most likely explanation for this small impact of inadequate management on the response to EBR/GZR is the low prevalence of EBR-related RASs observed in Spain.[17] Moreover, the SVR rates for genotype 1a observed in our study are in agreement with those reported in the C-EDGE treatment-naïve trial.[7]

Our results in former PWID without OAT are in agreement with those reported by the US Department of Veterans Affairs (VA) healthcare system.[18] In an analysis of the VA cohort, approximately half of the patients had a history of drug use. The rates of SVR for patients with history of drug use were similar to that of individuals without a prior drug use in the VA study.[18] A separate analysis of the VA healthcare system focused on patients with ICD codes of opioid use disorder or with OAT prescription showed high efficacy among them.[19] An SVR rate of 96% for those patients matches the overall SVR rates for EBR/GZR in the VA healthcare system.[18,19] However, in our study, only 91% of PWID on OAT achieved SVR. Moreover, recent drug users showed sharply lower responses. Barely 70% recent drug users reached SVR. Recent drug use was not analysed in the VA studies.[18,19] Thus, we provide new information on the particularly hard to treat group of PWID with active drug use.

Our study has a number of strengths. First, we report the efficacy and safety of EBR/GZR in patients included in real-world multicenter cohorts. Second, nearly two thirds of the study population were comprised by PWID with and without OAT. The overall efficacy of EBR/GZR in PWID on OAT found in the C-EDGE CO-STAR trial was replicated in this real-world sample of patients.[8,20] However, this study may have certain limitations. First, recent drug use was retrospectively assessed by clinical records and chart reviews, where self-reported data are registered and not by scheduled urine testing. Due to this, we most likely underestimated recent drug use. Second, the frequency of genotype 1a was higher among PWID than patients who never injected drugs, who were mainly infected by genotype 1b. HCV genotype-related differences in response to EBR/GZR could underlie the lower rates of SVR among PWID. After excluding patients discontinuing therapy because of non-treatment-related reasons, in the mFAS analysis, we found similarly high rates of SVR by genotype. In addition, after adjusting by recent drug use, HCV genotype was not associated with the likelihood of SVR in the multivariate analysis. Third, only part of the relapses was evaluated by phylogenetic analysis. Among PWID with recent drug use, we might have misclassified reinfections as relapses. However, all three relapses among recent drug users were assessed, and reinfections were ruled out on the basis of phylogenetic analysis. Finally, active alcohol intake could have influenced the adherence of patients to EBR/GZR. Unfortunately, the use of alcohol was not collected for the study patients.

In conclusion, EBR/GZR is effective in PWID with and without OAT. However, recent drug use increases the likelihood of voluntary drop-out and, as a consequence, to low response rates to EBR/GZR. Spain is on track to meet the 2030 HCV elimination targets by WHO. However, the proportionally small group of active drug PWID is a serious barrier to reach the elimination of HCV as a public health threat. Specific strategies designed to increase the retention of recent drug users are needed to make the elimination of HCV by 2030 an attainable goal.