The Effects of Sustained Virological Response to Direct-Acting Anti-Viral Therapy on the Risk of Extrahepatic Manifestations of Hepatitis C Infection

Hashem B. El-Serag; Israel C. Christie; Amy Puenpatom; Diana Castillo; Fasiha Kanwal; Jennifer R. Kramer


Aliment Pharmacol Ther. 2019;49(11):1442-1447. 

In This Article


We estimated the risk of developing several possible HCV-related extrahepatic manifestations in a large cohort of DAA-treated patients and analysed the effect of SVR on the extrahepatic condition risk. We found that successful DAA treatment resulting in SVR was associated with significant reductions in the future risk of several conditions, including mixed cryoglobulinaemia, glomerulonephritis and lichen planus but not NHL or diabetes. The magnitude of risk reductions ranged between 0.23 and 0.61, and the findings were robust for several adjustments for potential confounders, including several demographic and clinical variables, additional risk factors for these extrahepatic manifestations and healthcare utilisation. There was risk reduction in porphyria cutanea tarda; however, even in our large cohort, the incidence of porphyria cutanea tarda was too low to allow for a suitably powered statistical analysis.

Our study is the first to address the possible effect of DAA-related SVR on the risk of developing new cases of extrahepatic manifestation rather than the resolution of existing ones. The findings highlight the causal role of active HCV in these conditions and the role of DAA-related SVR as a preventive measure. The findings also support growing literature about successful DAA treatment and resolution of mixed cryoglobulinaemia vasculitis with[11,12] or without renal[13] involvement with minimum side effects[14] and a mechanism related to changes in peripheral B- and T-cell homoeostasis.[15] A previous study with a similar design conducted using national VA data among patients receiving interferon-based HCV therapy reported risk reductions when patients with SVR were compared with treated patients without SVR for mixed cryoglobulinaemia, glomerulonephritis, porphyria cutanea tarda and diabetes but not NHL.[4]

We did not show statistically significant reduction in the incidence of porphyria cutanea tarda, NHL or diabetes in the DAA-related SVR group compared to the DAA treated, no-SVR group. Studies have consistently shown both strong association between HCV and porphyria cutanea tarda and resolution of porphyria cutanea tarda, with successful HCV treatment including DAAs.[16,17] However, in our study, the incidence of new porphyria cutanea tarda cases in the SVR group was 19 (4 in the no-SVR group), and, although the difference in the numbers is clear and in favour of SVR, it was not statistically significant due to low power. Diabetes has been variably associated with active HCV infection,[18] and some but not all reports indicate a possible improvement in glycaemic control and HgbA1c and decreased inulin use[19] following DAA-related SVR.[20–22] However, the attributable fraction to HCV is small, and therefore, the expected reduction in diabetes risk with SVR is expectedly very small. The link between chronic HCV infection and a subset of B-cell NHL is strongly supported by epidemiological studies.[23] Peginterferon/ribavirin therapy has proved its crucial role in the cure of these NHLs, and DAA-related SVR has been associated with a favourable response among patients with B-cell lymphoproliferative disorders, including NHLs[24] and possibly marginal zone lymphoma.[25,26] However, B-NHL is a heterogeneous group of lymphomas with varied clinical presentation and may be indolent or aggressive. Anti-viral treatment may be sufficient for low-grade lymphomas, but chemotherapy is necessary in patients with high-grade lymphomas.[27]

The results of this study should be considered in the context of several limitations. First, since this is an observational study, selection bias is possible. However, the study was limited to those who received DAA treatment, thus minimising the possible differences between the SVR and non-SVR groups. Also, we only examined the first course of DAA treatment. This may have resulted in some patients who were in the no-SVR group eventually having an SVR; however, since our follow-up time was relatively short, we expect this occurrence only in few people with minimal effect on the results. Second, patients with a shorter duration of follow-up time are less likely to obtain incident diagnoses, and this is independent of their SVR status but an artefact of the length of follow-up in the database. However, the duration of follow-up was not appreciably different by SVR status. Third, some of the incident extrahepatic manifestation diagnoses may represent prevalent or preexisting cases; however, this misclassification is likely to be similar among the SVR and no-SVR groups. Furthermore, all risk estimates were adjusted for healthcare utilisation, as measured by the number of outpatient visits, a surrogate measure of the likelihood of making new diagnoses. Fourth, our definition for the extrahepatic manifestations might have low specificity and falsely identified additional patients with the conditions of interest; therefore, we created a more restrictive specific definition and found comparable results. In addition, our SVR4 definition may have overidentified patients with SVR; therefore, we conducted a sensitivity analysis using SVR12 and found comparable results. Finally, our results may not be generalisable outside the VA healthcare system as our cohort was overwhelmingly male, older and lower socioeconomic status than HCV patients outside the VA system.

In addition to its unique focus on the preventive effect of DAA on extrahepatic manifestations, our study has several other strengths. The large sample size coupled with moderately long duration of follow-up (average 2 years) allowed capture of sufficient outcome events in both those with and without SVR to allow a meaningful analysis of most conditions of interest. The study also uses accurate, laboratory test-based definitions of HCV infection and SVR as well as DAA treatment based on pharmacy records. The observed overall SVR of 92.1% is consistent with effectiveness data for DAA in community-based studies and further adds to the internal validity of the results.[28,29]

The study also provides unique estimates of the prevalence and incidence of several possible extrahepatic manifestations in patients with HCV, stratified by SVR status. The prevalence of these extrahepatic conditions in this study is 7.09% and 35.46% if diabetes is considered. The risk attributable to HCV in these conditions is unclear but is likely to be high in porphyria cutanea tarda; low in diabetes; and intermediate for NHL, lichen planus and glomerulonephritis. Successful DAA treatment resulting in SVR was associated with significant reductions in the risk of several conditions, including mixed cryoglobulinaemia; glomerulonephritis lichen planus; and, possibly, porphyria cutanea tarda but not non-Hodgkin lymphoma or diabetes. These data should be considered when discussing overall burden of HCV disease and cost-effectiveness of screening and treatment.