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

Disclosures

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

In This Article

Results

Study Selection

A total of 9077 citations were identified through an initial search, among which, 3901 duplicated records were removed. Further 5042 records were excluded after title and abstract screening. The full text of the remaining 134 records were reviewed, and 31 eligible records were retained. In addition, 23 records were added from reference lists of included records and updated search. Ultimately, 54 records with 56 eligible studies were included in the analysis (one record consists of three independent studies, Figure 1).

Study Characteristics

Study characteristics were summarised in Table 1 and Table 2. Fifty-six studies[11,13,28,30,31,36–84] were included, with a total of 5522 patients with HCC, while 27 studies enrolled non-HCC patients as control groups. Fifty-three (95%) studies were observational cohorts (45 retrospective and eight prospective) and three studies were clinical trials. In 40 studies (75%), 3910 patients all received curative HCC management, in five studies (9%), 983 patients received either curative or noncurative HCC management and in nine studies (16%), 361 patients received noncurative HCC management. The proportion of patients with advanced cirrhosis was not reported in 31 studies (55%), was ≥50% patients in four studies (7%) and <50% of patients in 21 studies (38%).

Risk of Bias Assessment

The results of the risk of bias assessment within included studies are shown in Table S3. Eighteen studies were below score 3, 28 studies scored at 4–5 and 10 studies had score 6–7.

Synthesis of Results

In 56 studies included in the analysis (5522 patients with HCC), the pooled ITT SVR was 88.3% (95% CI 86.1–90.4), with a high heterogeneity across studies (I 2 = 85.0%, P < .001). Estimated pooled mITT SVR, available among 45 studies including 4069 patients, was 89.6% (95% CI 87.4–91.8, I 2 = 83.0%, P < .001), comparable with ITT SVR. In 27 studies enrolling both patients with prior or present HCC (n = 3126) and patients without HCC (n = 49 138), pooled SVR was 88.2% (95% CI 85.0–91.4, I 2 = 86.5%, P < .001) in the HCC population and 92.4% (95% CI 91.1–93.7, I 2 = 96.1%, P < .001) in the non-HCC population. Pooled OR of SVR among HCC population compared to no HCC population was 0.54 (95% CI 0.43–0.68, I 2 = 57.4%, P < .001), indicating significantly lower SVR among HCC population compared to non-HCC population. Similar findings were observed, restricting the analysis to 11 studies only including patients with cirrhosis (Table 3, Figure 2).

Figure 2.

Comparison of sustained virologic response (SVR) between hepatocellular carcinoma (HCC) patients and non-HCC patients (A) and between HCC and non-HCC patients, all with cirrhosis (B)

Stratified Analysis

In 42 studies with all patients receiving curative HCC management (4178 patients), pooled SVR was 90.4% (95% CI 88.3–92.4, I 2 = 81.1%, P < .001). In five studies of 490 patients receiving liver transplantation, pooled SVR was 95.4% (95% CI 92.4–98.4, I 2 = 53.6, P = .13). A lower SVR was seen in five studies (983 patients) with either curative or noncurative HCC management (pooled SVR 78.9%, 95% CI 76.4–81.5, I 2 = 0, P = .9) and in nine studies (361 patients) with noncurative HCC (pooled SVR 82.5%, 95% CI 73.9–91.2, I 2 = 83.7%, P < .001). High heterogeneity was observed across studies in each category (Table 4, Figure 3).

Figure 3.

Forest plots of sustained virologic response (SVR) in all hepatocellular carcinoma (HCC) patients (A), in patients with curative HCC management (B), in patients with mixed HCC management (C) and in patients with noncurative HCC management (D)

Fourteen studies reported SVR by DAA regimen. Pooled SVR was 76.3% (95% CI 60.4–92.1, I 2 = 90.1%, P < .001) in 208 patients who received suboptimal DAA therapy, 89.2% (95% CI 83.3–95.1, I 2 = 71.7, P < .001) in 356 patients who received PI-containing DAA therapy and 96.9% (95% CI 94.3–99.4, I 2 = 54.3, P = .01) in 856 patients who received SOF and NS5A inhibitor DAA therapy (Table 5, Figure 4).

Figure 4.

Sustained virologic response (SVR) to suboptimal direct-acting antivirals (DAAs) (A), protease inhibitor-containing DAAs (B) and sofosbuvir + NS5A DAAs (C)

Fifteen studies reported SVR by HCV genotype. Pooled SVR was 92.0% (95% CI 88.1–95.6, I 2 = 71.3%, P < .001) among 1087 patients with HCV genotype 1, 89.3% (95% CI 82.7–96.0, I 2 = 0, P = .51) among 90 patients with HCV genotype 2 and 77.1% (95% CI 51.4–100, I 2 = 87.5, P < .001) among 93 patients with HCV genotype 3 (Table 6, Figure 5).

Figure 5.

Sustained virologic response (SVR) in patients with hepatitis C virus (HCV) genotype 1 (A), genotype 2 (B) and genotype 3 (C)

Twelve studies (1799 patients) reported post-DAA liver function status. Significant improvement of liver function was seen in 95.3% of patients (nine studies, 1715 patients), while 3.5% of patients (two studies, 62 patients) had nonsignificant improvement and 1.2% of patients (one study, 22 patients) had no worsening of liver function. The improvement of liver function included declines of AFP, MELD score, Child-Pugh score and liver stiffness.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....