IFNL4 Genotype is Associated with Virologic Relapse After 8-Week Treatment with Sofosbuvir, Velpatasvir and Voxilaprevir

To the Editor:

Direct acting anti-viral agents (DAAs) have revolutionized the treatment of chronic infection with hepatitis C virus (HCV), but further improvements are needed to meet the public health goal of controlling HCV worldwide. Shortened courses of therapy might reduce the high costs that present a major barrier to large scale treatment and improve options for treating HCV in marginalized, high prevalence populations with potentially poorer adherence to treatment. In developing intervention strategies, it is important to differentiate virologic failure (most often due to virologic relapse) from poor adherence as these problems require different responses. Previously, we analyzed published data and demonstrated that genotype for the IFNL4 rs12979860 polymorphism (often called ‘IL28B’) was significantly associated with virologic relapse after 8-week therapy with ledipasvir/sofosbuvir among HCV genotype 1-infected treatment naïve patients without cirrhosis enrolled in the ION-3 trial; relapse occurred in only 1.8% of patients with the favorable IFNL4 rs12979860-CC genotype compared to 5.8% of patients with the IFNL4 rs12979860-CT or TT.^{1, 2} We demonstrated a similar relationship between IFNL4 genotype and relapse among patients treated in 12-week trials of ledipasvir/sofosbuvir.¹ IFNL4 rs12979860 is a genetic marker for the functional IFNL4-ΔG/TT (rs368234815) polymorphism,³ which is the primary genetic determinant of HCV clearance. These polymorphisms provide generally similar results, however, IFNL4-ΔG/TT is the better predictor in patients of African ancestry.³ Studies of IFNL4 genotype and response to interferon-based therapies demonstrated that the lower frequency of favorable IFNL4 alleles found among African Americans explained racial differences in spontaneous HCV clearance and response to interferon-α treatment.^3–5 Jacobson et al now report results for POLARIS-2, a phase III clinical trial of sofosbuvir/velpatasvir/voxilaprevir in fixed dose combination.⁶ We performed additional analyses of those published data to examine whether IFNL4 rs12979860 genotype predicted relapse following 8 weeks of treatment with this pan-genotypic regimen.

Among the 501 patients randomized to 8 weeks of sofosbuvir/velpatasvir/voxilaprevir in POLARIS-2, 21 (4%) had virologic relapse and none had virologic breakthrough.⁶ Per our previous approach, we calculated relapse rates among patient subgroups using the frequency of baseline characteristics to determine denominators and supplemental descriptive information for patients who relapsed to determine numerators. On that basis, only 2/166 (1.2%) patients with IFNL4 rs12979860-CC genotype relapsed compared to 15/253 (5.9%) for IFNL4 rs12979860-CT and 4/82 (4.9%) for IFNL4 rs12979860-TT. Therefore, patients enrolled in POLARIS-2 who had an unfavorable IFNL4 rs12979860 genotype (CT or TT) were about five times more likely to relapse than those with IFNL4 rs12979860-CC (relative risk of virological relapse, 4.76; p=0.02, Fischer exact test). In other recent reports, it has been suggested that African American race may be an independent risk factor for poorer response to sofosbuvir-based therapy,^{7, 8} but that relationship is confounded by the lower frequency of favorable IFNL4 genotypes among black individuals.^3–5 In the 8-week arm of POLARIS-2, relapse was more common in blacks (6.3%) than non-blacks (4.0%), however, only 9.6% of all study subjects in that arm were black and the difference in relapse by race did not approach statistical significance (p=0.44). Therefore, the association between unfavorable IFNL4 rs12979860 genotype and relapse among patients receiving 8 weeks of sofosbuvir/velpatasvir/voxilaprevir cannot be explained by race.

The observation that IFNL4 genotype predicts virologic relapse after shortened regimens of sofosbuvir-based therapy has potential implications for the treatment of individual patients and for national strategies to control HCV. In wealthier countries with higher treatment costs, IFNL4 genotype might be determined for individual patients and the appropriate duration of therapy could be based on IFNL4 genotype and other variables. In some lower resource countries, generic DAAs may be affordable, but individualized IFNL4 genotyping is cost prohibitive. In such settings, national treatment strategies might consider the overall frequency of favorable IFNL4 genotypes in that population. For example, compared to individuals of European ancestry, Asians are much more likely to carry favorable IFNL4 genotypes (rs12979860-CC and rs368234815-TT/TT), therefore, 8-week treatment with sofosbuvir/velpatasvir/voxilaprevir might yield considerably lower rates of virologic relapse in Asian populations than was seen in POLARIS-2.

In summary, our analysis of data from POLARIS-2 extends previous observations that IFNL4 genotype is associated with virologic relapse in response to sofosbuvir-based regimens for chronic hepatitis C. Higher rates of failure with sofosbuvir-based treatment among black patients compared to whites reported in some studies likely reflect a lower frequency of favorable IFNL4 genotypes in that population. On the other hand, given that Asians have the highest frequency of favorable IFNL4 genotypes, relapse may be most infrequent in Asian populations. Consideration of these relationships might improve global strategies to treat and control HCV infection.