Treating Hepatitis C in an Era of Interferon-free Therapies
Treating Hepatitis C in an Era of Interferon-free Therapies
With the continuation of Scotland's existing treatment strategy (i.e. S1000-SQ), liver-related deaths increase from 130 cases in 2015, up to 150 cases in 2020, and stabilise thereafter (see figure 1). New instances of SLM peak at 195 cases in the year 2015 (i.e. the year we assume new treatments will be introduced), then declines continuously to 145 cases in 2030. Similarly, incident chronic infections fall steadily from 565 cases in 2015, to 440 cases by 2030.
(Enlarge Image)
Figure 1.
Modelled number of incident (I) severe liver morbidity (SLM), (II) liver deaths and (III) chronic infections, through treating 1000 persons per year, according to various distribution strategies. IFN, interferon.
Reducing Incident Cases of Severe Liver Morbidity. Strategies prioritising persons with moderate and advanced fibrosis (i.e. S1000-ADV and S2000-ADV) exerted the optimal impact on SLM. Over the long-term time horizon (i.e. 2015–2030) cumulative SLM cases are 36.1% lower (95% CI −41.4 to −23.4) with S2000-ADV, than under S1000-SQ. PWID-focused strategies also reduce SLM, but to a lesser extent; e.g. analogously, S2000-PWID leads to a 26.2% reduction (95% CI −32.3 to −14.7) (see figures 1–3 and Table 4). A no-prioritisation approach increases the number of SLM cases, relative to a status quo case-mix. In fact, in the long term, treating 2000 patients a year without prioritisation had an equivalent impact on SLM as treating half that number under a fibrosis prioritising strategy. That is to say, incident SLM was 11.4% lower (95% CI −17.9 to +27.7) under S2000-NP, but 12.8% lower (95% CI −19.8 to −6.8) with S1000-ADV. Notably, SLM does not decline indefinitely towards zero. For example, even under our maximum S2000-ADV strategy, SLM stabilises at 70 cases per year. Of these 70 cases, the majority (40–45 cases per year) arise in patients who have previously attained SVR.
(Enlarge Image)
Figure 2.
Modelled number of incident (I) severe liver morbidity (SLM), (II) liver deaths and (III) chronic infections, through treating 2000 persons per year, according to various distribution strategies. IFN, interferon.
(Enlarge Image)
Figure 3.
Treatment strategy performance: %-change in cumulative number of incident: (I) severe liver morbidity (SLM), (II) liver deaths and (III) chronic infections, between 2015–2030, relative to S1000-SQ.
Reducing Liver Mortality. Similarly, strategies targeting those with moderate and advanced fibrosis have the greatest impact on curtailing liver mortality, while a no-prioritisation approach increased liver deaths (relative to S1000-SQ) (see figures 1–3 and Table 4). Thus, performance with respect to liver mortality was a mirror image of performance with respect to SLM. A key difference between SLM and liver mortality, however, is that the latter is less amenable to change than the former. Particularly, in the short term, where even with the S2000-ADV strategy, only a 2.4% mortality reduction (95% CI −5.0 to −1.5) is seen relative to S1000-SQ. Nevertheless, important differences do emerge thereafter; e.g. over the longer term, a 21.4% mortality reduction (95% CI −24.9 to −13.4) occurs with S2000-ADV, relative to S1000-SQ. (but this decline remains notably less than the 36.1% reduction in SLM seen under the same S2000-ADV strategy and over the same time period).
Reducing Incident Chronic Infections. PWID-focused strategies bring about rapid steadfast declines in incident infection (see figures 1–3 and Table 4). Over the long term, S2000-PWID leads to a 52.2% (95% CI −55.9 to −23.9) reduction in incident infections relative to S1000-SQ. This performance far exceeds rival strategies— i.e. analogously, only a 13.3%–20.4% reduction occurred with S2000-ADV, S2000-SQ and S2000-NP. Of further mention, under S2000-PWID, the important public health goal of reducing incident chronic infection to negligible levels (i.e. to <50 cases per year) is achieved in 2025.
Combination Strategies: (see online supplementary eTables7 and 8 and eFigures 2 and 3 http://gut.bmj.com/content/64/11/1800/suppl/DC1). Under the S2000-ADV strategy, Scotland evidently begins to run out of patients with advanced fibrosis from the year 2025 onwards (see online supplementary eTable 10 http://gut.bmj.com/content/64/11/1800/suppl/DC1). The consequence then is that in these latter years, less than the 2000 persons are treated per annum in our model under this strategy. Similarly, with the S2000-PWID strategy, Scotland runs out of infected PWID from ~2025, and so, overall, <2000 patients are treated each year. Given this, we created a posthoc combination strategy, whereby we first target those with advanced fibrosis in 2015–2022, and then change our focus to PWID in 2023–2030 (we also consider the vice-versa strategy). This combination approach offers an improved compromise between minimising both liver-related complications and new chronic infections over the 2015–2030 timeframe.
Rapid Devastation of SLM: Rapid devastation of SLM (see online supplementary eTable 10 http://gut.bmj.com/content/64/11/1800/suppl/DC1). The S2000-ADV strategy stabilises SLM at 70–75 cases by the year 2023 (i.e. within our medium-term time horizon). We explored, in a posthoc interrogation of our model, how this same reduction in SLM might be expedited. More specifically, we posed the following question: How many patients would Scotland need to treat under a moderate/advanced fibrosis prioritising strategy, in order to effect a rapid devastation of SLM? We defined a 'rapid devastation' as reducing incident SLM to <75 cases within our short-term time horizon (2015–2020). On that basis, the minimum such treatment uptake necessary to achieve this goal was 3250 patients per year (i.e. a 3.25-fold increase from existing uptake levels, which amounts to treating 1980 individuals with moderate/advanced fibrosis per year).
Results
Disease Forecasts Assuming Continuation of Existing s1000-sq Strategy
With the continuation of Scotland's existing treatment strategy (i.e. S1000-SQ), liver-related deaths increase from 130 cases in 2015, up to 150 cases in 2020, and stabilise thereafter (see figure 1). New instances of SLM peak at 195 cases in the year 2015 (i.e. the year we assume new treatments will be introduced), then declines continuously to 145 cases in 2030. Similarly, incident chronic infections fall steadily from 565 cases in 2015, to 440 cases by 2030.
(Enlarge Image)
Figure 1.
Modelled number of incident (I) severe liver morbidity (SLM), (II) liver deaths and (III) chronic infections, through treating 1000 persons per year, according to various distribution strategies. IFN, interferon.
Performance of Alternative Treatment Strategies (Versus Existing S1000-SQ Strategy)
Reducing Incident Cases of Severe Liver Morbidity. Strategies prioritising persons with moderate and advanced fibrosis (i.e. S1000-ADV and S2000-ADV) exerted the optimal impact on SLM. Over the long-term time horizon (i.e. 2015–2030) cumulative SLM cases are 36.1% lower (95% CI −41.4 to −23.4) with S2000-ADV, than under S1000-SQ. PWID-focused strategies also reduce SLM, but to a lesser extent; e.g. analogously, S2000-PWID leads to a 26.2% reduction (95% CI −32.3 to −14.7) (see figures 1–3 and Table 4). A no-prioritisation approach increases the number of SLM cases, relative to a status quo case-mix. In fact, in the long term, treating 2000 patients a year without prioritisation had an equivalent impact on SLM as treating half that number under a fibrosis prioritising strategy. That is to say, incident SLM was 11.4% lower (95% CI −17.9 to +27.7) under S2000-NP, but 12.8% lower (95% CI −19.8 to −6.8) with S1000-ADV. Notably, SLM does not decline indefinitely towards zero. For example, even under our maximum S2000-ADV strategy, SLM stabilises at 70 cases per year. Of these 70 cases, the majority (40–45 cases per year) arise in patients who have previously attained SVR.
(Enlarge Image)
Figure 2.
Modelled number of incident (I) severe liver morbidity (SLM), (II) liver deaths and (III) chronic infections, through treating 2000 persons per year, according to various distribution strategies. IFN, interferon.
(Enlarge Image)
Figure 3.
Treatment strategy performance: %-change in cumulative number of incident: (I) severe liver morbidity (SLM), (II) liver deaths and (III) chronic infections, between 2015–2030, relative to S1000-SQ.
Reducing Liver Mortality. Similarly, strategies targeting those with moderate and advanced fibrosis have the greatest impact on curtailing liver mortality, while a no-prioritisation approach increased liver deaths (relative to S1000-SQ) (see figures 1–3 and Table 4). Thus, performance with respect to liver mortality was a mirror image of performance with respect to SLM. A key difference between SLM and liver mortality, however, is that the latter is less amenable to change than the former. Particularly, in the short term, where even with the S2000-ADV strategy, only a 2.4% mortality reduction (95% CI −5.0 to −1.5) is seen relative to S1000-SQ. Nevertheless, important differences do emerge thereafter; e.g. over the longer term, a 21.4% mortality reduction (95% CI −24.9 to −13.4) occurs with S2000-ADV, relative to S1000-SQ. (but this decline remains notably less than the 36.1% reduction in SLM seen under the same S2000-ADV strategy and over the same time period).
Reducing Incident Chronic Infections. PWID-focused strategies bring about rapid steadfast declines in incident infection (see figures 1–3 and Table 4). Over the long term, S2000-PWID leads to a 52.2% (95% CI −55.9 to −23.9) reduction in incident infections relative to S1000-SQ. This performance far exceeds rival strategies— i.e. analogously, only a 13.3%–20.4% reduction occurred with S2000-ADV, S2000-SQ and S2000-NP. Of further mention, under S2000-PWID, the important public health goal of reducing incident chronic infection to negligible levels (i.e. to <50 cases per year) is achieved in 2025.
Posthoc Analyses
Combination Strategies: (see online supplementary eTables7 and 8 and eFigures 2 and 3 http://gut.bmj.com/content/64/11/1800/suppl/DC1). Under the S2000-ADV strategy, Scotland evidently begins to run out of patients with advanced fibrosis from the year 2025 onwards (see online supplementary eTable 10 http://gut.bmj.com/content/64/11/1800/suppl/DC1). The consequence then is that in these latter years, less than the 2000 persons are treated per annum in our model under this strategy. Similarly, with the S2000-PWID strategy, Scotland runs out of infected PWID from ~2025, and so, overall, <2000 patients are treated each year. Given this, we created a posthoc combination strategy, whereby we first target those with advanced fibrosis in 2015–2022, and then change our focus to PWID in 2023–2030 (we also consider the vice-versa strategy). This combination approach offers an improved compromise between minimising both liver-related complications and new chronic infections over the 2015–2030 timeframe.
Rapid Devastation of SLM: Rapid devastation of SLM (see online supplementary eTable 10 http://gut.bmj.com/content/64/11/1800/suppl/DC1). The S2000-ADV strategy stabilises SLM at 70–75 cases by the year 2023 (i.e. within our medium-term time horizon). We explored, in a posthoc interrogation of our model, how this same reduction in SLM might be expedited. More specifically, we posed the following question: How many patients would Scotland need to treat under a moderate/advanced fibrosis prioritising strategy, in order to effect a rapid devastation of SLM? We defined a 'rapid devastation' as reducing incident SLM to <75 cases within our short-term time horizon (2015–2020). On that basis, the minimum such treatment uptake necessary to achieve this goal was 3250 patients per year (i.e. a 3.25-fold increase from existing uptake levels, which amounts to treating 1980 individuals with moderate/advanced fibrosis per year).
Source...