Improvement of ALT decay kinetics by all-oral HCV treatment: Role of NS5A inhibitors and differences with IFN-based regimens

Background Intracellular HCV-RNA reduction is a proposed mechanism of action of direct-acting antivirals (DAAs), alternative to hepatocytes elimination by pegylated-interferon plus ribavirin (PR). We modeled ALT and HCV-RNA kinetics in cirrhotic patients treated with currently-used all-DAA combinations to evaluate their mode of action and cytotoxicity compared with telaprevir (TVR)+PR. Study design Mathematical modeling of ALT and HCV-RNA kinetics was performed in 111 HCV-1 cirrhotic patients, 81 treated with all-DAA regimens and 30 with TVR+PR. Kinetic-models and Cox-analysis were used to assess determinants of ALT-decay and normalization. Results HCV-RNA kinetics was biphasic, reflecting a mean effectiveness in blocking viral production >99.8%. The first-phase of viral-decline was faster in patients receiving NS5A-inhibitors compared to TVR+PR or sofosbuvir+simeprevir (p<0.001), reflecting higher efficacy in blocking assembly/secretion. The second-phase, noted δ and attributed to infected-cell loss, was faster in patients receiving TVR+PR or sofosbuvir+simeprevir compared to NS5A-inhibitors (0.27 vs 0.21 d-1, respectively, p = 0.0012). In contrast the rate of ALT-normalization, noted λ, was slower in patients receiving TVR+PR or sofosbuvir+simeprevir compared to NS5A-inhibitors (0.17 vs 0.27 d-1, respectively, p<0.001). There was no significant association between the second-phase of viral-decline and ALT normalization rate and, for a given level of viral reduction, ALT-normalization was more profound in patients receiving DAA, and NS5A in particular, than TVR+PR. Conclusions Our data support a process of HCV-clearance by all-DAA regimens potentiated by NS5A-inhibitor, and less relying upon hepatocyte death than IFN-containing regimens. This may underline a process of “cell-cure” by DAAs, leading to a fast improvement of liver homeostasis.


Introduction
In 1998 Neumann and his colleagues proposed a seminal model to explain the biphasic decline of HCV-RNA after initiation of interferon (IFN) [1]. According to this model, the rapid viralload decline in the first days of treatment depends upon virus clearance from plasma, with a magnitude reflecting the treatment effectiveness in blocking viral production. After this rapid "first-phase", a slower"second-phase" ensues, consistent with progressive elimination of infected hepatocytes.
In the last 20 years, this initial model was expanded to include novel ideas [2][3][4][5], and novel drugs. One important point raised by scientists was that the second-phase of viral kinetics during direct-acting antivirals (DAAs) treatment may not only represent cell-elimination, but may also be associated with the progressive reduction of the intracellular viral content, down to its disappearance (i.e. "cell-cure") [6][7][8]. This possibility was supported by experiments in vitro [9] and by the clinical observation that the second-phase with DAAs was faster than during IFN-based therapy, and no increase in Alanine Amino Transferases (ALT) levels was observed [7,10].
ALT levels can be considered as a surrogate marker for cell-death, since hepatic injury causes their leak from damaged hepatocytes and their increase in serum [11]. Since IFN-activity includes cell-death, it was not uncommon to observed suboptimal ALT decline, and infrequent ALT-normalization, during IFN-treatment, especially in patients with more compromised liver status [12].
Consistently with the idea of a "cell-cure" by DAAs, ALT kinetics is thus supposed to be improved. Telaprevir (TVR) indeed allowed ALT-normalization in 62% of patients by day-9 when used in monotherapy (vs. 33% when used with pegIFN and ribavirin [PR]) [13], and the IFN-free association of SOF+ledipasvir+GS9669/GS-9451 allowed ALT reduction to normal levels by day-14 in 90% of patients [14]. Nevertheless, no data are available on ALT kinetics with currently-used DAA combinations in clinics, nor on the impact of cirrhosis, making difficult to appreciate the role of all-DAA combinations in the more complex patients seen in realworld practice.
Here, in order to better understand the effect of DAA combinations, we compared the kinetics of both ALT and HCV-RNA in cirrhotic-patients treated with various DAA-based regimens, with or without IFN, for whom frequent measurements of both markers were available. We used the techniques of mathematical modeling to recapitulate the kinetics observed and better understand the origin of possible differences across treatment-regimens.

Study population
Adults aged !18 years with genotype-1 infection and cirrhosis, receiving one DAAs with or without IFN were included if had: a) baseline and week-4 HCV-RNA and ALT available; b) at least one HCV-RNA determination within the first 6 days of treatment, and one HCV-RNA and ALT determination at week-1 and/or at week-2; c) high baseline ALT (>55 IU/ml in men; >45 IU/ml in women). Patients who prematurely ended treatment before week-4 were excluded.
Cirrhosis was documented by liver-biopsy or Fibroscan (!12.5 KPa). Detailed information on sampling and laboratory analyses are reported in S1 Supplementary Material.
Approval by the local Ethics Committees of Fondazione PTV Policlinico Tor Vergata and patient written informed consent were obtained. This study was conducted in accordance with the principles of the Declaration of Helsinki. All information were recorded in an anonymous database.

Statistical analysis
Results were compared using Spearman's rank correlation test, Mann-Whitney U-test, Fisher exact test or Chi 2 test.
Cox regression analysis was performed using the following confounding variables: IFNadministration, HCV-subtype, baseline ALT, NS5A-inhibitors administration, HCV-RNA decay baseline-2 weeks, baseline HCV-RNA, previous IFN-experience (relapse, no-response, other), previous PI-failure, IL-28B CC genotype, and ribavirin co-administration. Only variables having a p-value <0.10 in univariate analysis were included in multivariable analysis. To avoid statistical multicollinearity, IFN and NS5A-administration were analyzed separately. Statistical significance threshold was set at 0.050.
All statistical analyses were performed using the open source environment R and SPSS software package (SPSS Inc., Chicago, Illinois).

Mathematical modeling of early HCV-RNA and ALT kinetics
As we focused on early HCV-RNA and ALT kinetics, only data obtained during the first 4 weeks of treatment were analyzed. Data below LLOQ were taken into account using the extended SAEM algorithm implemented in MONOLIX 4.3.3.
HCV-RNA and ALT kinetics were analyzed separately using nonlinear mixed-effect models, optimized according to significant covariates as described in S1 Supplementary Material.
To illustrate the difference in the virological and biological response across the covariates, we simulated 1000 individual profiles from the final parameter estimates. On the best model for viral and ALT kinetics containing only significant covariates, the correlation between viral and ALT kinetic parameters, stratified or not on the different treatment groups, was studied using Spearman correlation test. A joint model that would simultaneously fit HCV-RNA and ALT kinetic was considered relevant if at least one correlation coefficient was >0.4.

Patterns of HCV-RNA and ALT decay during and after antiviral treatment
In the first 4 weeks of treatment, a fast first-phase HCV-RNA decline in serum, followed by a slower second-phase, was observed in all 111 patients analyzed (Fig 1, panels A and B). No significant differences were found in HCV-RNA response rates among all-DAA and TVR+PR treated patients, at least until day-28 (Fig 1, panel C). On the contrary, and despite comparable baseline ALT values, TVR+PR patients showed a more heterogeneous, and generally suboptimal ALT decline compared to all-DAA patients (Fig 1, panels B). Indeed, all-DAAs treatment led to ALT normalization in 85.0% of patients after only 14 days, and in 93.8% after 28 days, while only 65.5% of TVR+PR patients had normal ALT at day-14 (p = 0.02), and 73.3% at day-28 (p = 0.001). The achievement of healthy ALT during treatment required a median (IQR) of 7 (7)(8)(9)(10)(11)(12)(13)(14) days for all-DAA patients, vs. 14 (14-39) in TVR+PR patients, even though this estimation can be affected by different sampling times.
After 12 weeks since treatment discontinuation, the proportion of patients with normal ALT values among those who had received IFN increased form 50% to 73.1%, vs. 91.9% of all-DAA patients (p<0.001) (Fig 1, panel D ALT normalization rates at EOT were not significantly affected by baseline RASs presence (p = 0.510), nor were SVR rates (p = 0.386), not even in patients with baseline HCV-RNA > 800.000 IU/ml (p = 0.152).

Impact of different DAA-combinations on modeled HCV-RNA kinetics
The kinetics of HCV-RNA decline was well described by a standard biphasic model, whose population-parameters are reported in Table 2. The mean effectiveness of antiviral treatment (ε) was 99.8% and 99.9% in patients treated with all-DAA or TVR+PR treatments, respectively (p = 0.01). Among the tested covariates (see Methods) NS5A-inhibitors significantly affected the clearance of free-virus (c) and the loss rate of infected-cells (δ). Indeed, patients receiving a NS5A-inhibitor had a faster rate of viralclearance (7.88 vs. 5.28 d -1 in NS5A-free; p<0.001), and a lower rate of loss of infected-cells (0.21 vs. 0.273 d -1 in NS5A-free; p = 0.0012). Thus, NS5A-inhibitors were associated with a more rapid first-phase decay of HCV-RNA, but a slower second-phase, as can be seen by the kinetics obtained by simulation in Fig 2,

Impact of different DAA-combinations on ALT kinetics
By Cox analysis, the use of IFN and of NS5A-free regimens were associated with higher risk of not achieving normal ALT values before EOT (both p<0.0001), even after normalization for several confounding variables (Table 3).
Notably, by mathematical modelling the rate of ALT decline (λ) was significantly lower in NS5A-free regimens than in NS5A-containing (0.17 vs. 0.27 d -1 , respectively; p<0.001), with no significant difference in TVR+PR patients and those receiving SOF+SIM (Table 2). However, on the long run, IFN-treated patients had a significantly higher ALT steady state (ALT ss ), than those receiving IFN-free regimens (43.4 vs. 26.3 IU/mL; p<0.001), regardless of NS5Ainhibitors administration. Typical patterns of ALT kinetics in the different treatment groups are shown in Fig 2,

Joint kinetics of HCV-RNA and ALT
Mathematical modeling showed no association between the viral kinetic and the ALT kinetics parameters (all correlation coefficients <0.4) in the overall population, including the loss-rate of infected-cell δ and the ALT normalization rate λ (S1 and S2 Figs). This remained true when analyzing separately patients receiving IFN+TVR and all-DAA (S1 and S2 Figs). Because the kinetics of viral decline was more profound in patients treated with IFN+TVR while the ALT decline was slower, it is not surprising to find that, for a given level of viral-load reduction, the corresponding level of ALT decline is much larger in patients treated with IFN-free regimen than in patients treated with IFN+TVR (Fig 2, panels A and B). This is also illustrated in Fig 3, where the predicted levels of ALT and HCV-RNA at different time-points are represented. For none of these time-points, the ALT decline levels were found correlated with HCV-RNA decline, even when data were analyzed separately for each treatment group (not shown). This, in other words, suggests that the decline of ALT and HCV-RNA are not reflecting similar mechanisms, and that these mechanisms may be different in IFN-treated and all-DAA.

Discussion
To our knowledge, this is the first study analyzing and comparing the different kinetics of HCV-RNA and ALT during all-DAA treatments, unveiling the effect of a specific DAA-class. We focused our study on GT-1 patients with cirrhosis, a critical population of patients with peculiar physio-pathological and anatomical characteristics, poorly studied for viral kinetics.
Our mathematical inference predicted a markedly faster and sharper ALT kinetic during all-DAA treatment than with TVR+PR, as well as a lower ALT steady-state at week-4.
In the context of IFN-based therapy, modeling papers tend to support a correlation between HCV-RNA and ALT kinetics [4], though with large inter-patients variability according to liver status [12]. This correlation was explained by IFN use, that acts by establish a non-virus-  specific antiviral state [15], ultimately leading to a progressive clearance of serum HCV-RNA mediated by both clearance of circulating virions (first-phase) and, afterwards, by the death or clearance of productively-infected hepatocytes (second-phase) [1], eventually leading to serum ALT increase. Indeed, previous kinetics studies in IFN-treated patients showed a positive correlation between lower infected-cell death (δ) and persistently normal ALT values [16]. The typical ALT decay expected in patients with elevated ALT during effective IFN-treatment would thus be a superposition of two trends: 1) an early ALT decline due to initial viral clearance; and 2) a slower increase during the first 4-8 treatment weeks due to IFN-activity, 3) reversible after EOT in sustained virologic responders. We indeed observed this fluctuating ALT decay during TVR+PR treatment, when the initial ALT decay was followed by possible re-alteration and, lastly, by ALT normalization after EOT.
Notably, after achieving SVR, 73.1% of TVR+PR patients had normal ALT values, a significantly lower proportion respecting to all-DAAs treated patients (92%), but still higher of the 67% observed in cirrhotic patients receiving only IFN [12], supporting a further activity of TVR decay and ALT normalization. Indeed, recently DAAs, and TVR in primis, were proposed by kinetic studies to act by progressive reduction of intracellular viral-RNA down to its disappearance ("cell-cure"), rather than by inducing cell-death [7,8]. The results of our study support this hypothesis, by demonstrating a biphasic ALT decline in which only ALT decline and normalization were observed: "healthy" ALT levels were always achieved during all-DAA treatment before complete HCV-RNA clearance, and then maintained until EOT.
The positive effects of all-DAA regimens on ALT kinetics was not simply an effect of a more profound HCV-RNA decay, and, consequently, of a sharper reduction in HCV-related pro-inflammatory status [17]. Indeed, the improved ALT decline at each analyzed time point during all-DAA treatments was confirmed also after normalization for HCV-RNA decay, again supporting an independent process of cell-cure by DAAs. By analyzing different all-DAA combinations, we demonstrated, for the first time to our knowledge, a significant enhancement of ALT declines by NS5A-inhibitors. The use of this DAA-class also improved the concomitant first-phase of viral kinetics, in agreement with previous studies that demonstrated a rapid inhibition of viral assembly and secretion by NS5Ainhibitors [6]. NS5A-inhibitors may thus simultaneously potentiate early HCV-RNA decay by accelerating viral clearance rate from serum [6], and "anticipate" the process of cell-cure by more rapid intracellular inhibition of HCV life-cycle, even though this last hypothesis needs to be confirmed by in vitro studies, not yet available. Notably, both these positive effects were lost during second-phase, when δ was low and the predicted ALT kinetic profile was no longer NS5A-dependent.NS5A-inhibitors are now a universal backbone for first-line DAA regimens [18,19], and their proposed ability in improving cell-cure further supports their preferential use as initial treatment for chronic or acute HCV infection, along with their high antiviral activity and safety profile.
Our study has some points of potential weakness. It does not include a control-population of patients receiving only PR, though TVR has a poor effect on vRNA kinetics [20], and data on "pure IFN" treatments are widely available in literature as historical controls. In addition, ALT values at early time points, such as 48h and week-1 of treatment, were available only for few TVR+PR treated patients, making the Cox analysis less precise during the first days.
Another point worth to be considered is that we analyzed a population of cirrhotic patients, in whom liver architecture is altered, and both HCV-RNA and ALT decays can be affected [12,21,22]. Even if our results and conclusions may be not directly extended to non-cirrhotic patients, the presence of cirrhosis seemed not to affect glaringly ALT decline. In a recent study with SOF+ledipasvir+GS9669/GS-9451, including mostly non-cirrhotic patients, ALT normalization was achieved in 90% of patients by day-14 [14], vs. 85% of our all-DAA patients. Differently from ALT, viral kinetics was generally reduced in our study. Our free-virus clearance rate value is lower than previous estimates (c<8 vs. c = 22.3 day -1 ) [6], maybe as consequence of advances cirrhosis and/or few early sampling-points. Second-phase was also quite slow and associated with poor response rates at week-4, concordantly with previous results in cirrhotic patients [23]. Whether the absence of correlation between second phase of viral decline and the normalization rate of ALT also holds in non-cirrhotic patients will need to be examined.
Overall treatment effectiveness ε was 0.998, matching previous results obtained in non-cirrhotic patients [7,24,25]. However, viral-kinetic response under treatment is not necessarily correlated with treatment outcome [26], and indeed TVR+PR patients achieved SVR 12 only in 66.7% of cases, vs. 93.8% in all-DAAs. In the latter population, SVR 12 rates were reduced in CTP-B patients (88.9%) compared to CTP-A (95.6%), concordant with previous results from the SOLAR study [27], and more recent data on SOF/DCV in ALLY-1 trial [28], and SOF/velpatasvir combination [29]. These results probably depend upon the severe liver alteration in decompensated patients, that deeply affect blood perfusion and metabolic capacity of the residual liver-tissue, making "cell cure" by DAAs much more difficult. Indeed, in our CTP-B patients also ALT kinetics was particularly slow, and 12.5% of CTP-B patients reached EOT with still altered ALT levels vs. only 1.6% of CTP-A patients; p = 0.08.
In summary, we provided a detailed analysis of HCV-RNA and ALT kinetics profiles in cirrhotic GT-1 patients treated with different DAAs-based regimens, with particular attention to the effect of NS5A-inhibitors use and highlighting differences with previous IFN-containing regimens. Our results uncovered a strikingly fast ALT normalization in patients receiving NS5A-inhibitors, in contrast with a persistent ALT alteration in patients receiving IFN, independent by HCV-RNA clearance. Considering serum ALT levels as a surrogate parameters of hepatocyte turnover [30, 31], the DAA-driven "sparing" and curing of liver cells (differently by the cytopathic effect of IFN) may provide an intriguing rationale for these results.