Exclusion criteria in studies of direct-acting antivirals (DAAs) for hepatitis C (HCV) infection would have kept high proportions of HCV/HIV-coinfected individuals out of those studies, according to analysis of eligibility for five trials by members of the Canadian Coinfection Cohort. The most frequent exclusion criteria ruled out certain antiretroviral combinations or active illicit drug use.
DAAs revolutionized treatment of HCV infection, typically curing more than 90% of participants in controlled trials. Sustained virologic response (SVR) rates have also been high in HCV patients coinfected with HIV. But because most trials exclude certain subgroups, it remains unclear whether these good results with DAAs apply to most coinfected people with HIV infection.
To address that question, researchers working with the Canadian Coinfection Cohort determined proportions of cohort members who would have been excluded from five large DAA efficacy trials that enrolled HCV/HIV-coinfected patients and tested diverse DAAs: NCT01479868 (simeprevir [TMC435]), PHOTON-1 (sofosbuvir [Sovaldi]), TURQUOISE-I (ombitasvir, paritaprevir/ritonavir/dasabuvir [Viekira Pak]), ION-4 (sofosbuvir/ledipasvir [Harvoni]) and ALLY-2 (daclatasvir [Daklinza] + sofosbuvir [Sovaldi]). Up to April 1, 2015, the Canadian Coinfection Cohort had enrolled 1423 coinfected people at 18 centers. After excluding people who died or dropped out of the cohort and people with genotypes not reflecting trial populations, the researchers had 615 patients with evidence of chronic HCV infection.
The target study population had a median age of 49 years and 28% were women. A high proportion, 80%, had a history of injection drug use, and 30% currently injected drugs. High proportions in this cohort were below the poverty level (73%) or had a history of incarceration (56%). Yet most people (86%) were taking combination antiretroviral therapy, and most (78%) had an undetectable HIV load.
Trial eligibility criteria would have excluded all but 5.9% of cohort members (24 of 410) from NCT01479868, 9.8% (52 of 530) from PHOTON-1, 6.3% (26 or 410) from TURQUOISE-I and 8.1% (34 or 421) from ION-4. In contrast, 43% of cohort members (233 of 541) would have been eligible for the more inclusive ALLY-2 trial.
In the four more exclusionary trials, the most frequent exclusion criteria were restriction to specific antiretrovirals (excluding from 63% to 79% of the Canadian Coinfection Cohort) and active illicit drug use (excluding 53% to 55%). Even if cohort members could have safely switched to acceptable antiretroviral regimens, the four more exclusionary trials would have barred 74% to 77% from screening, largely because of active illicit drug use. Less frequent exclusion criteria included detectable HIV RNA (15% to 18% excluded) and minimal CD4+ counts (3% to 19% excluded).
Because these five DAA trials would have excluded most HCV/HIV-coinfected people in the Canadian cohort, the researchers conclude that these "trial results may have limited generalizability" to a wider HCV/HIV population. They determined "that the majority of exclusionary criteria were not related to safety but appear to be aimed at maximizing treatment response rates." The authors argue that exclusion of active drug users "may have been overly conservative as studies have shown they can achieve comparable SVRs as those not injecting drugs in well-supported settings." They call for observational studies to gauge DAA treatment response in coinfected populations "to determine how effective these therapies will be in the real world."