French hepatitis C care cascade: substantial impact of direct-acting antivirals, but the road to elimination is still long

With the advent of highly effective direct-acting antivirals (DAAs) for hepatitis C virus (HCV) infection, in 2014, the World Health Assembly pledged to eliminate HCV infection as a public health threat, through reducing new infections and deaths by 90 and 65%, respectively, by 2030 [1]. To achieve this, in 2016, the World Health Organization (WHO) launched a global strategy which aimed to diagnose 90% of people with HCV chronic infection and treat 80% of eligible patients by 2030 [2].

To monitor progress towards the WHO’s goal, evaluating the cascade of care (CoC) – a widely used tool for HIV infection [3] – may also be effective for HCV infection. This consists in estimating the numbers or percentages of infected people at various stages of the care pathway, from diagnosis to treatment. Identifying gaps in the CoC can help guide Public health policy. Although numerous studies have evaluated the HCV CoC using prospective or retrospective cohorts (i.e., a longitudinal approach), their samples comprised captive, often local, populations, including people attending screening or clinical centres [4, 5], which limits extrapolation of their results to the wider population. Evaluating the HCV CoC at a national level is essential, but challenging. It is often based on modelled data [6, 7] or on combining existing national data sometimes covering a large period of time [8‐10]. However, the latter approach is not adapted to the rapidly evolving situation in the era of DAAs. Furthermore, published studies on the HCV CoC in the DAA era are scarce [7, 11, 12].

France has been involved for many years in the fight against hepatitis C, in particular through three national plans implemented since the end of the 1990s [13]. This has led to a slightly more favourable epidemiological situation than in most countries [14, 15], with an estimated viremic infection prevalence in mainland France in 2004 of 0.53% (95% Confidence Interval (CI): 0.40–0.70), corresponding to 232,000 chronically-infected individuals (95% CI: 167,869–296,523) in the 18–80 years old (y.o.) general population, and an estimated 57% (95% CI: 41–71) of these people being aware of their infection, corresponding to 132,000 (95% CI: 95,000-165,000) individuals [16]. Since then, HCV screening activity has continuously increased, reaching 62 anti-HCV tests per 1000 inhabitants in 2016 (vs. 53 / 1000 in 2010) [17], despite no change in official screening recommendations for targeted people at risk of exposure. With regard to antiviral treatment, 1st and 2nd-wave DAAs have been rapidly and widely implemented in France, since 2011 and 2014, respectively, despite their restricted access for people with severe liver disease until mid-2016 [18]. In this favourable context, France has committed itself to achieving HCV elimination by 2025 [19].

National evaluations of the CoC for a given year are needed to monitor progress towards HCV elimination and to assess the impact of DAAs. Accordingly, in this study, we aimed to evaluate the CoC for chronic HCV in mainland France in 2011 and in 2016, before and after the introduction of 2nd-wave DAAs.

We used a cross-sectional approach to estimate chronic HCV CoC in 2011 and in 2016.

The estimated numbers of people at the different stages of the chronic HCV CoC according to the main and age range-adjusted analyses, and their evolution between 2011 and 2016, are presented in Table 3. The corresponding CoC for both years are illustrated in Fig. 3.

This work provides the first estimate of the French national CoC for chronic hepatitis C for 2 years, before and after the introduction of 2nd-wave DAAs, in the context of the goal to eliminate HCV. Between 2011 and 2016, the number of people with HCV chronic infection decreased by 30%, the proportion of people aware of their infection substantially increased (from 57.7 to 80.6%), and the numbers of people receiving care for HCV infection and on antiviral treatment during the year both increased by one quarter.

This work has several strengths: first, it focused on the French population at the national level (i.e., not local or specific populations), enabling us to assess French progress toward the WHO’s elimination objective [2]. Second, given the rapidly evolving situation in recent years, with the advent of DAAs in recent years, it provided estimates of the different stages of the CoC related to the same year (whether 2011 or 2016), not a larger period [8‐10]. Third, almost all estimates of the numbers of people at the different stages were independently calculated. Estimates in some previous studies of national CoC [8, 10, 29] were interdependent, which could pose problems in the event of an error in one of the estimates. Fourth, SNDS data are not subject to selection bias because they cover almost the entire French population. Finally, by estimating the HCV CoC before (2011) and after (2016) the introduction of 2nd-wave DAAs, we were able to assess these drugs’ initial impact on the CoC.

Our work also has limitations: first, different methodologies were used to estimate the numbers of people with chronic HCV infection and those aware of their infection, specifically, modelling studies for 2011 [22, 23] and a cross-sectional survey (BaroTest) in the general population for 2016 [24]. In the latter type of survey, some populations at risk of HCV (e.g., active injecting drug users) are frequently not represented or are underrepresented, leading to possible underestimation of national prevalence. The difference we observed between 2011 and 2016 prevalence estimates is therefore probably overestimated. Second, given the low prevalence in the general population, the number of persons testing positive in BaroTest was small, leading to poor robustness in the estimated proportion of people aware of their infection in 2016 [24]. Third, because SNDS does not include biological test results or data on consultation diagnoses, we had to construct algorithms to identify people receiving HCV care. Validation studies would be necessary to assess these algorithms’ performance. Fourth, the identification of people on antiviral treatment depends on coding quality, which was suboptimal for DAAs in 2014–2015 but less so in 2016 [18, 30]. Finally, because of data sources, our CoC could not consider people older than 80 y.o. or those living in French overseas areas and could not be presented according to age, gender, at-risk exposures and time from infection (to focus on new or re-infected people).

The decrease in chronic hepatitis C prevalence (from 0.42% in 2011 to 0.30% in 2016) is probably mostly linked to highly effective 2nd-wave DAAs which led to at least 20,000 people in France being cured in 2014–2015 [18]. A decrease in prevalence (from 1 to 0.6%) was also observed in a population-level CoC in British Columbia between 2012 and 2018 [11]. This trend can also be explained by deaths and aging in the HCV population beyond the age limit of the prevalence surveys (80 years for 2011, 75 years for 2016). Indeed, in France, a large percentage of HCV contaminations occurred before the implementation of systematic testing of blood donors and harm reduction measures in the 1990s, leading to a rather elderly population of people with chronic HCV infection. A continuation of the decrease in the incidence among people who actively inject drugs - observed between 2004 and 2011 (from 15.4 to 11.2 per 100 person-years) [31] - is another possible explanation.

The increasing trend in the proportion of chronically infected people aware of their infection observed in our study is coherent with other published estimates for France [6, 15, 32]. This trend may be partly explained by the substantial increase in anti-HCV screening activity in laboratories (+ 21% between 2010 and 2016) [17], which in turn is possibly related to the availability of DAAs and to experts’ recommendations in 2014 and 2016 advocating universal screening [14, 33]. In addition, the decrease in the number of elderly infected patients (whether through death or aging beyond the age limit of related studies), which constituted the age-group with the highest proportion of undiagnosed cases [23], may have contributed to the large observed increase in the proportion of people aware of their infection between 2011 and 2016.

The numbers of people receiving care for chronic HCV infection and on antiviral treatment during the year increased by 22.5 and 24.6%, respectively, between 2011 and 2016. As depicted in Additional file 4, these numbers evolved in parallel with the advent of new antiviral drugs.

Our results show a substantial improvement in all the stages of the chronic HCV CoC over a short 5-year period, characterized by major therapeutic innovations, specifically the advent of 1st and 2nd-wave DAAs, which were rapidly and widely implemented in France [18].

In 2016, according to our estimates, among people with chronic HCV infection, more than eight in ten were aware of their infection, one in four received care for their HCV infection, and one in eight were on antiviral treatment during the year. Our results are relatively consistent with 2015 Markov model-based estimates by the European Union HCV Collaborators group which found that among chronically-infected people in France (estimated prevalence = 0.29%), 74.1% were diagnosed and 10.2% started treatment [6]. Moreover, the group’s estimates for the European Union as a whole (estimated prevalence = 0.64%) [6], were 36.7 and 7.5%, respectively, which highlights France’s favourable situation when compared with other countries in the European Union (4th position for the diagnosis rate, and 2nd position for the proportion of persons treated) [6], which is even more favourable when compared with the rest of the world (20% - diagnosis and 1.5% - newly treated) [34]. Since 2016, these figures have placed France on the short list of countries considered to be on track to eliminate hepatitis C by 2030 [35]. A recent modelling study projected that France will reach HCV elimination by 2024 [36]. Indeed, France has already met the eight criteria identified in the study for successful HCV elimination, in particular the following essential criteria: i) political will, ii) a financed “priority prevention” plan (in 2018 as a result of criterion i)) which aimed to achieve HCV elimination by 2025 [19], iii) the removal of treatment restrictions in mid-2016 (which led to a 35% increase in the number of patients initiating a DAA treatment between 2016 and 2017 [30]), and iv) monitoring and evaluation of existing programs. The other criteria are: v) the expansion of treatment provision beyond specialists (DAAs can be delivered in retails pharmacies since March 2017 and prescribed by all physicians since May 2019), vi) the implementation of an awareness campaign in the general population (“Noise against the Hepatitis C”) [37], vii) a national screening program (annual local campaigns since 2019 and a national linkage to care program with outreach actions), and viii) the existence of harm-reduction programs (in particular, reinforced test and treat programs) [38].

Although it is expected that this comprehensive elimination program implemented in France in recent years will help enhance the country’s HCV CoC, our results demonstrate that in 2016, France was still far from reaching WHO’s elimination targets, in particular regarding the proportion of treated patients (only 12% versus the WHO objective of 80%) [2]. Our work highlights that care management for chronic HCV infection constitutes a major gap with an estimated 31.9% people receiving care among those aware of their infection in 2016. The choice of a conservative algorithm to identify people in care may partly explain this low proportion, which should be considered as a minimum. However, the application of the alternative, more sensitive algorithm led to a slightly increased proportion (38.2%). Universal access to DAAs in France (announced in mid-2016 but only effective since mid-2017), may further increase this proportion. The diagnosis of infected people also constitutes a challenge for HCV elimination. Experts and patient associations strongly advocated universal screening of all adults at least once during their lifetime, in addition to risk-based targeted testing [14, 33], after a modelling study showed that this new strategy could be cost-effective [39]. However, in 2019, the French National Authority for Health concluded that its effectiveness was not sufficiently demonstrated [40].

Our study demonstrates a marked improvement in the CoC for chronic HCV infection between 2011 and 2016, suggesting the substantial impact of 2nd-wave DAAs in mainland France. However, access to care and antiviral treatment remained insufficient in 2016, when considering the WHO’s elimination target of 2030. These national estimates will have to be updated to assess the impact of new measures implemented since 2016, and to monitor the elimination of HCV by 2025, as planned by France. Further studies are also needed to estimate the CoC among specific populations including people who inject drugs.