Determining the Lower Limit of Detection Required for HCV Viral Load Assay for Test of Cure Following Direct-acting Antiviral-based Treatment Regimens

Evidence From a Global Data Set

Jake R. Morgan; Elizabeth Marsh; Alexandra Savinkina; Sonjelle Shilton; Shaun Shadaker; Tengiz Tsertsvadze; George Kamkamidze; Maia Alkhazashvili; Timothy Morgan; Pam Belperio; Lisa Backus; Waheed Doss; Gamal Esmat; Mohamed Hassany; Aisha Elsharkawy; Wafaa Elakel; Mai Mehrez; Graham R. Foster; Constance Wose Kinge; Kara W. Chew; Charles S. Chasela; Ian M. Sanne; Yin M. Thanung; Anne Loarec; Khawar Aslam; Suna Balkan; Philippa J. Easterbrook; Benjamin P. Linas


J Viral Hepat. 2022;29(6):474-486. 

In This Article


Characteristics of Treatment Failures

Overall, our cohort consisted of 5973 cases of detectable viraemia following HCV treatment from Egypt (3264), the United States (1125), Georgia (1041), the United Kingdom (131), Myanmar (84), Cambodia (40), Pakistan (27), Ukraine (17), Mozambique (3) and clinical trial registries (241). Table 2 summarizes the characteristics of the individuals included in this analysis. Most individuals were under the age of 60 years (65%), with the exception of the U.S. Veterans Affairs cohort, of which 78% were 60 years or older. The majority of patients in the Mozambique, Myanmar and Ukraine cohorts were HIV co-infected, but less than 10% were HIV co-infected in the other cohorts and trial databases. Forty-one per cent of the cohort had a fibrosis stage of F4 indicating advanced liver disease/cirrhosis. The genotype distribution across all cohorts was GT 1 (23%), 2 (5%), 3 (12%) and 4 (53%). Most of the cohort (70%) received 12 weeks of DAA treatment while 24% received 24 weeks. Most individuals received a mid-era DAA regimen (66%), followed by early era (27%) and recent-era (6%) DAAs (Table 2).

HCV Viral Load Distribution and Limit of Detection Analysis

The median detectable HCV RNA at SVR12 was 287,986 IU/mL (approximately 5.5 log10) (IQR = 1,323,500 IU/mL with 25th percentile = 26,500 and 75% percentile = 1,350,000) (Figure 1). 90% of those with detectable viraemia at SV12 had a viral load greater than 1133 (95% CI 940–1390), 95% greater than 227 IU/mL (95% CI 170–276), 97% greater than 70 IU/mL (95% CI 48–86) and 99% greater than 19 IU/mL (95% CI 16–23). Five hundred seventy-four individuals (10%) were defined as having LLV, meaning that a hypothetical assay with LLoD of 1000 IU/mL would miss approximately 10% of treatment failures in this setting.

Figure 1.

Each bar represents the proportion of the sample with a given log RNA value at time of detection 12 weeks post-treatment. The labels on the x-axis are the end point of each bar. For example, the tallest bar, with a label of 5.75–6.00, shows that just over 10% of the cohort has a log RNA value between 5.75 and 6.00

Factors Associated With Low-level Viraemia (<1000 IU/mL) at SVR12

We examined baseline demographic, clinical and treatment characteristics associated with LLV (<1000 IU/mL) compared to non-LLV and present these results in Table 3. In multivariable logistic regression analysis adjusting for demographic (age, sex, data source, HIV/HBV co-infection) and disease (fibrosis stage, genotype, regimen duration, DAA treatment era) characteristics, females had higher odds of experiencing LLV (odds ratio [OR] = 1.60, 95% CI 1.30–1.97). Compared to cirrhosis (F4), no or minimal fibrosis (F0-F1) was associated with higher odds of having a low detectable viral load (OR = 1.49, 95% CI 1.15–1.93), as was genotype 3 (OR = 1.69, 95% CI 1.18–2.41). Finally, we found that compared to early era DAA regimens, mid-era DAA regimens were associated with a lower likelihood of low-viraemia detection (OR = 0.55, 95% CI 0.40–0.75).

Sensitivity Analyses of Distribution of Viral Load

Clinical Trial and Observational Cohorts. We found a higher distribution of detectable viral loads from the two clinical trial registries. The median viral load was 2,344,229 (IQR = 5,911,542 IU/mL with 25th percentile = 545,000 and 75% percentile = 6,456,542), and 90% of those with detectable viraemia at SVR12 had a viral load greater than 98,420 (95% CI 17,600–199,962), 95% greater than 4030 IU/mL (95% CI 24–4100), 97% greater than 923 IU/mL (95% CI 24–4030) and 99% greater than 24 IU/mL (95% CI 14–24), respectively (Table 4). The distribution of viral load from non-pharmaceutical trials (observational databases) had median viral load of 264,809 (IQR = 1,196,500 IU/mL with 25th percentile = 23,500 and 75% percentile = 1,220,000), and 90% of those with detectable viraemia at SVR12 had a viral load greater than 1062 (95% CI 816–1300), 95% greater than 214 IU/mL (95% CI 166–266), 97% greater than 69 IU/mL (95% CI 48–85) and 99% greater than 19 (95% CI 16–22) (Table 4).

SVR12 and SVR24. We identified 432 individuals with 24 weeks post-treatment HCV RNA data, including 231 individuals from the U.S. Veterans Affairs cohort, 128 from the Georgia cohort and 73 from the clinical trials cohort. In this sample at SVR12, 95% of individuals had a detectable HCV RNA above 200 IU/mL, 97% above 119 IU/mL and 99% above 24 IU/mL. We graphed the 12- and 24-week viral load by individual in the 65 individuals in the clinical trials cohort with data at both time points (eFigure S2). In this sample, the median change in viral load from 12 to 24 weeks was 128,848 IU/mL. We did not find evidence of different prevalence of LLV at 12 and 24 weeks: at both the 12- and 24-week assessments for cure, approximately 11% of those detectable had LLV of <1000 IU/mL. So, while the mean increase from 12 to 24 weeks is large, this was largely attributable to increases in those with already high viral loads.