Effect of Hepatitis C Treatment with Ombitasvir/Paritaprevir/R + Dasabuvir on Renal, Cardiovascular and Metabolic Extrahepatic Manifestations: A Post-Hoc Analysis of Phase 3 Clinical Trials

Table 2 presents the baseline characteristics of the study population. Across the different clinical trials, the mean age of the population was 51 (50.1–57.7) years, with at least 69% (69–91%) of the population being in the F0–F2 fibrosis stages. With respect to clinical profile, at least 66% (66–83%) and 58% (58–74%) of the study population had BMI <30 kg/m² and HOMA-IR <3 mU mmol/l², respectively. There was fairly even distribution across sub-genotype 1a and 1b and at least 55% (55–82%) of patients were treatment naïve. SVR rates across the study population were greater than 95% (95–97.6%). Appendix Table S2 presents baseline characteristics of the patients by baseline EHM disease severity.

Descriptive analysis of the proportion of patients experiencing improvements showed across EHMs a greater proportion of patients who had severe EHM manifestations at baseline improved by greater than 20% from baseline (Table 3). These effects were sustained at all time points.

Regression analyses on cardiovascular EHMs show a significant reduction in mean triglyceride levels 12 weeks EOT with 3D ± RBV vs. placebo. The mean adjusted change from baseline to EOT week 12 for treated patients was −33.7 mg/dl (p < 0.0001; 95% CI −41.1, 26.5 mg/dl), while patients in the placebo group did not have statistically significant changes (−5.6 mg/dl at week 12; p = 0.53; 95% CI −23.5, 12.2 mg/dl) (Fig. 1a). Patients with elevated levels of triglycerides at baseline experienced greater declines in triglycerides during treatment compared to patients with normal triglycerides. Moreover, this difference persisted 52 weeks post treatment (Fig. 1b): patients with elevated triglycerides experienced a decline of 35.3 mg/dl (p < 0.001; 95% CI −48.2, −22.3 mg/dl), and patients with normal triglycerides had an increase of 16.2 mg/dl (p < 0.001; 95% CI 11, 21.4 mg/dl), which did not result in levels over 150 mg/dl.

Treatment with 3D ± RBV also significantly improved metabolic EHMs, as the adjusted mean glucose levels 12 weeks EOT were significantly lower in patients treated with 3D ± RBV vs. placebo [mean adjusted change of −12.1 mg/dl (p < 0.0001) for treated patients vs. −4.6 mg/dl (p = 0.02) for the placebo group; Fig. 2a]. Among the EHM subgroups, patients with pre-diabetes and diabetes experienced greater reduction in glucose levels during treatment (Fig. 2b) compared to patients with normal glucose levels. This difference persisted 52 weeks post treatment. Patients with pre-diabetes and diabetes experienced a decline of 4.4 mg/dl (p = 0.004; 95% CI −7.3, −1.4 mg/dl) and 34.2 mg/dl (p < 0.0001; 95% CI −39.4, −28.9 mg/dl), respectively, and patients with normal glucose had a statistically insignificant increase of 1.9 mg/dl (p = 0.062; 95% CI −0.09, 3.85 mg/dl), which did not result in levels over 100 mg/dl (Fig. 2b).

Renal EHMs were not significantly affected after 12 weeks of treatment with 3D ± RBV vs. placebo as both groups had no statistically significant changes in eGFR values from baseline [mean adjusted change of +0.37 ml/min/1.73 m² (p = 0.82) for patients treated with 3D ± RBV vs. +3.48 ml/min/1.73 m² (p = 0.19) for patients in placebo arm] (Fig. 3a). However, the subgroup of patients with CKD stage 2 and 3 experienced minor to significant improvements in eGFR during treatment (Fig. 3b) compared to patients with normal eGFR values, respectively. By 52 weeks post treatment, stage 3 patients experienced an improvement of 1.6 ml/min/1.73 m² (p = 0.7; 95% CI −6.6, 9.9 ml/min/1.73 m²). Stage 2 patients experienced a statistically insignificant decline of 2.2 ml/min/1.73 m² (p = 0.13; 95% CI −4.9, 0.6). Stage 1 patients experienced a decrement of 9.3 ml/min/1.73 m² (p < 0.001; 95% CI −10.8, −7.8 ml/min/1.73 m²) (Fig. 3b).The CKD stage 4 and 5 patients experienced no statistically significant decrements during treatment and returned to baseline values by week 12 post treatment (Fig. 3c). Stage 4: −0.81 ml/min/1.73 m²; p = 0.53; 95% CI −3.4, 1.7 ml/min/1.73 m². Stage 5: 0.09 ml/min/1.73 m²; p = 0.9; 95% CI −1.25, 1.43 ml/min/1.73 m².

Unadjusted analysis is presented in appendix Table S3 for the three EHMs and is consistent with the adjusted results presented.

The mean decrement of triglycerides in treated patients at 52 weeks post treatment was 9.5 mg/dl overall and 35.2 mg/dl for patients with elevated triglycerides at baseline. The magnitude of these declines is comparable to the improvements observed in clinical trials of triglyceride-lowering drugs, which ranged from 8 to 50% [34]. Elevated serum triglycerides are a known risk factor for coronary heart disease (CHD) and long-term all-cause mortality [18, 19], suggesting that treatment with 3D ± RBV may lead to potential long-term beneficial cardiovascular outcomes.

Fasting blood glucose levels over 110 mg/dl have been associated with vascular complications [20]. Similarly, recurrence of cardiovascular events have been observed to increase as fasting plasma glucose levels rise above 90 mg/dl and double with fasting plasma glucose levels of 110–115 mg/dl [21]. The American College of Endocrinology guidelines thus recommend a glycemic control target of <110 mg/dl [35]. Results from our analysis showed that the overall mean decrease in serum glucose levels at 52 weeks post treatment with 3D ± RBV was 12.3 mg/dl overall, 4.3 mg/dl in patients with pre-diabetes and 34.2 mg/dl in patient with diabetes. Moreover, these patients were more likely to experience a decrease in serum glucose levels of >20%, which increases the likelihood of reducing glucose levels below 110 mg/dl.

The results of this analysis showed no significant impairment of renal function with 3D ± RBV treatment, consistent with previous studies of 3D clinical trial data [36]. Our study did identify a decrease in mean eGFR at week 52 for patients with CKD stage 1, which does not appear to be clinically meaningful and was not seen in patients with more advanced CKD at baseline. This finding might be attributed to differences in baseline characteristics between CKD groups, since a higher proportion of stage 1 patients had BMI >30 and HOMA-IR score >3 at baseline as compared to patients with lower eGFR levels as baseline. These factors have previously been associated with a decline in eGFR [37‐39]. The overall decline in eGFR observed in treated patients during the post-treatment period is also consistent with the effect of aging as reported previously [40, 41]. A meta-analysis of 35 cohorts of patients enrolled in a CKD prognosis consortium concluded that the adjusted HR of all-cause mortality and end-stage renal disease (ESRD) was largely unchanged in patients with an increase or decrease in eGFR of 10% or less relative to patients with stable eGFR, but was increased with eGFR declines greater than 10%. Furthermore, patients with an increase in eGFR >10% had a lower risk of progression to ESRD [20]. In the current analysis more than 40% of the HCV-treated population who had a baseline eGFR <60 ml/min/1.73 m² experienced an eGFR improvement of up to 20%, suggesting that treatment with 3D ± RBV may reduce the risk of ESRD development and all-cause mortality in the long run.

Our findings should be interpreted within certain limitations. First, the parameters selected as surrogates for extrahepatic disease were chosen in a post hoc manner since they were routinely measured in clinical trials. If de novo trials were being conducted, different tests and biomarkers could improve the measurement of EHMs. For example, a fasting lipid panel would provide a more robust indication of cardiovascular risk than fasting triglycerides alone; serial HOMA-IR calculations would more precisely measure changes in glucose metabolism; serial urine collection would have allowed a more detailed description of the impact of treatment on renal disease. In addition, since the analysis was conducted on patients enrolled in clinical trials, it may have limited generalizability to the overall population. However, results seem consistent with prior literature assessing different patient populations. Furthermore, our analyses by baseline EHM severity did not distinguish between patients who did or did not achieve SVR. Nonetheless, it is likely that antiviral efficacy drives improvement in EHM biomarkers as the overall SVR rate in these trials exceeded 95% and our regression models found a statistically significant association between viral load reduction and EHM biomarker improvements. This study might also suffer from unobservable bias where clinical variables (e.g., other comorbidities or concomitant medications) not collected in the database might influence the study results. To mitigate this issue, we leveraged double-blind trial data on patients randomly assigned to HCV treatment or placebo to assess the effect of treatment on EHM outcomes. Additionally, a real-world study did not find concomitant medications to be a significant predictor of EHM clinical outcomes in the presence of antiviral treatment [15]. In addition, we extrapolated the improvement observed in EHM biomarkers to clinical outcomes based on prior published literature. This may not be accurate and may warrant further research using real-world data with confirmed diagnosis of clinical outcomes. Finally, it should be noted that some of the patients included in the analysis did not receive an approved regimen as indicated in the current product label.

Nevertheless, this analysis contributes to the understanding of HCV treatment outcomes. It is one of the first to assess the impact of an oral IFN-free DAA regimen on EHM outcomes. It leverages a broad base of clinical trial data and included a comparison between treatment and placebo arms to evaluate the effect of DAA therapy on EHM outcomes.