Systematic Review With Meta-analysis

Effectiveness of Direct-acting Antiviral Treatment for Hepatitis C in Patients With Hepatocellular Carcinoma

Sichan He; Ian Lockart; Maryam Alavi; Mark Danta; Behzad Hajarizadeh; Gregory J. Dore


Aliment Pharmacol Ther. 2019;51(1):34-52. 

In This Article


Protocol and Registration

This study was conducted and reported in accordance with Preferred Reporting System for Systematic Review and Meta-analysis (PRISMA) statement[32] (Table S4). A study protocol was developed and registered with International Prospective Register of Systematic Review (PROSPERO) on 02 November 2018 (registration number CRD42018112222).

Eligibility Criteria

Studies investigating the response to DAA therapy among patients with HCV infection and prior or present HCC were included if they met all the following criteria:

  1. All or some study population included patients with prior or present HCC (studies with <10 patients with HCC were excluded; studies with only recurrent HCC patients were excluded);

  2. DAAs as HCV treatment (studies with small proportion of IFN-containing regimens with suboptimal DAAs were discussed by the study team for potential inclusion);

  3. Sustained virologic response (SVR) was reported among patients with HCC;

  4. Published in English.

Information Sources

A computer-aided search was conducted in bibliographic databases including PubMed, Web of Science, Scopus and Cochrane Central Register of Controlled Trials (CENTRAL). Presentations at the major viral hepatitis conferences were also searched, including International Liver Congress, and The Liver Meeting. The was searched for registered clinical trials. Reference lists of relevant studies were manually searched. Given that DAA therapy had been available since 2013, we searched the studies published since January 2013. Searches were conducted on 31 July 2018 and updated on 30 January 2019. Overall, studies published between 01 January 2013 and 30 January 2019 were covered in our search. A third updated search was performed in August 2019 covering a recently published systematic review[33] to identify eligible papers.

Search Strategy

Various terms related to HCV, HCC and DAA therapy were selected as search terms. Details of search strategy are documented in Table S1. Search strategies were developed based on PICO (Population, Intervention, Comparison and Outcome) framework, whereas no search terms were included for comparison and outcome to increase the sensitivity of the search strategies.

Study Selection

The details of the study selection process are provided in Figure 1. The records from databases through computer-assisted searches and hand searches were combined in an electronic EndNote library, while duplicated records were identified and removed. The records were initially screened by title and abstract. The full text of remaining records was further reviewed, with eligible studies selected for inclusion. Title and abstract screening was conducted by one author, while 10% of the records were screened by another author for verification. Full-text screening was conducted by two authors independently, with discrepancies discussed with the study team for a consensus decision.

Figure 1.

Review process and study selection. *One included record consists of three independent studies

Data Collection Process and Data Items

Required data were extracted from included studies to a pre-designed excel sheet for data extraction. The extracted data included study-level characteristics, patient-level characteristics and outcome variables (Table 1). Authors were contacted if the key variables were not reported.

Risk of Bias Assessment

Risk of bias of individual studies was assessed using a modified scale (Table S2) derived from Newcastle-Ottawa Scale for cohort studies.[34] The scale consists of five questions with a total score of 7, evaluating definition and representativeness of the study population, clear reporting of HCC management and outcome and definition and assessment of HCV treatment outcome. Risk of bias assessment was conducted by two authors independently, with discrepancies discussed with a third author to reach consensus.

Synthesis of Results

All statistical analyses were conducted using STATA Version 14.2 (StataCorp). The primary study outcome was SVR, defined as an HCV RNA level less than the lower limit of quantitation (or undetectable) at 12 or 24 weeks after the end of HCV treatment. The secondary study outcome was odds ratio (OR) of SVR among people with HCC compared to those with no HCC in studies that included both HCC and non-HCC patients. For each study, SVR was calculated based on intention-to-treat (ITT) and modified intention-to-treat (mITT) analyses. In mITT analysis, patients who died before SVR assessment were excluded, to explore whether HCC-related deaths impacted on SVR estimate. Heterogeneity of SVR and OR across studies was assessed using I-square (I 2) and Cochran's Q statistics. I 2 >50% indicates substantial heterogeneity. P < .10 for Cochran's Q statistics was considered as statistically significant.

Meta-analysis was used to pool the estimates, using random effects model (DerSimonian-Laird method). In meta-analysis of SVR, metaprop STATA package was used, and standard errors were calculated using the Wilson method.[35] A continuity correction of 0.5 was applied to the values in studies with 100% SVR. In meta-analysis of ORs, metan STATA package was used, and log-transformed ORs and standard errors were used in the analyses.

Stratified analysis was conducted by HCC management (ie curative vs mixed vs noncurative HCC management), DAA regimen (ie protease inhibitor containing vs sofosbuvir plus NS5A inhibitor vs suboptimal) and HCV genotype. Curative HCC management was defined based on complete response (CR) to HCC management. For the studies not reporting CR, liver transplantation, surgical resection and local ablation were considered as curative while transarterial chemoembolisation (TACE), external radiotherapy, transarterial radioembolisation (TARE) and systemic chemotherapy including sorafenib were considered as noncurative HCC management. TACE used to downstage HCC before curative management was considered a curative-intent modality. Suboptimal DAA regimens refer to sofosbuvir (SOF) and ribavirin (RBV), with or without interferon. Protease inhibitor (PI) containing DAAs refer to sofosbuvir/simeprevir (SOF/SIM), paritaprevir/ritonavir-ombitasvir (PrO or 2D regimen), paritaprevir/ritonavir-ombitasvir and dasabuvir (PrOD or 3D regimens), daclatasvir/asunaprevir (DCV/ASV), simeprevir/daclatasvir (SIM/DCV) and grazoprevir/elbasvir (GZR/EBR), with or without RBV. Sofosbuvir and non-nucleoside nonstructural protein 5A (NS5A) inhibitor containing DAA therapy refers to sofosbuvir/ledipasvir (SOF + LDV), sofosbuvir/daclatasvir (SOF/DCV), with or without RBV.

Advanced cirrhosis was summarised, based on diagnosis of decompensated cirrhosis, Child-Pugh Class B or C or a model for end-stage liver disease (MELD) Score ≥15. Interval from HCC treatment to DAA therapy was categorised by time from first/last HCC treatment to the initiation of DAA therapy. Time of last CR was alternatively recorded, if time of first/last HCC treatment was not available.