| |
|
Impact of Sustained Virologic Response with Direct-Acting Antiviral Treatment on Mortality in Patients with Advanced Liver Disease
|
| |
| |
Download the PDF here
Hepatology July 2017 - Lisa I. Backus MD PhD, Pamela S. Belperio PharmD, Troy A. Shahoumian PhD and Larry A. Mole PharmD Department of Veterans Affairs, Population Health Services, Palo Alto Health Care System, Palo Alto, CA, USA
"Our data indicated that the time to development of HCC was delayed in those who achieved SVR compared to those without SVR. Moreover, the risk of all-cause mortality was reduced in patients with a history of HCC who achieved SVR after DAA treatment. While the reduction in death rate was less pronounced in those with a history of HCC within 5 years of DAA initiation compared to those with no HCC history, achievement of SVR was still associated with a 64% reduction in death rate compared to those patients with no SVR, highly suggesting that successful treatment with DAAs in the setting of prior HCC was beneficial. This magnitude of reduction was similar to that observed in other smaller trials.(24,25,27)"
Abstract
The impact of sustained virologic response (SVR) on mortality after direct-acting antiviral (DAA) treatment is not well documented. This study evaluated the impact of DAA-induced SVR on all-cause mortality and on incident hepatocellular carcinoma (HCC) in 15,059 HCV-infected patients with advanced liver disease defined by a FIB-4 >3.25. Overall, 1,067 patients did not achieve SVR (No SVR) and 13,992 patients achieved SVR. In a mean follow-up period of approximately 1.6 years, 195 No SVR patients and 598 SVR patients died. Mortality rates were 12.3 deaths/100 patient years of follow-up for No SVR patients and 2.6 deaths/100 patient years for SVR patients, a 78.9% reduction (p<0.001).
Among patients without a prior diagnosis of HCC, 140 No SVR patients and 397 SVR patients were diagnosed with incident HCC. HCC rates were 11.5 HCC/100 patient years for No SVR patients and 1.9 HCC/100 patient years for SVR patients, an 83.5% reduction (p<0.001).
In multivariable Cox proportional hazard models controlling for baseline demographics, clinical characteristics and comorbidities, SVR was independently associated with reduced risk of death compared to No SVR (hazard ratio (HR) 0.26, 95% confidence interval (CI) 0.22-0.31, p<0.001). A history of decompensated liver disease (HR 1.57, 95%CI 1.34-1.83, p<0.001) and decreased albumin (HR 2.70 per 1 g/dL decrease 95%CI 2.38-3.12, p<0.001) were independently associated with increased risk of death. Conclusion: Those achieving SVR after DAA treatment had significantly lower all-cause mortality and lower incident HCC rates than those who did not achieve SVR.
DISCUSSION
In this evaluation, the high SVR rates attained with oral DAAs translated into an all-cause mortality benefit after adjusting for numerous baseline patient characteristics and comorbidities. In this large cohort of patients with advanced liver disease, achievement of SVR conferred a significantly reduced risk of death regardless of genotype for the entire cohort, for the subset of patients with prior HCC and for the subset of patients with prior hepatic decompensation. Furthermore, attainment of SVR reduced the risk of developing a first episode of HCC after DAA treatment by 84%.
The magnitude of risk reduction observed soon after achieving SVR underscores the importance of timely treatment in HCV-infected patients with advanced liver disease. The risk of mortality was 79% lower in those achieving SVR. From a population health perspective, as more patients are treated with potent all-oral DAA regimens and consequently more patients achieve SVR, more deaths will be averted than had been previously averted with older, less effective therapies. While the risk of all-cause mortality with peginterferon-based treatment was similarly reduced on the magnitude of 50%-75% in patients with SVR, overall fewer patients achieved SVR thus the population impact was less.(1,2)
The mechanism of the observed reduction in all-cause mortality associated with SVR cannot be elucidated with the present research but is undoubtedly multifactorial. We did not have information about reported cause of death so we cannot address what proportion of the reduction in mortality was potentially related to reduction in liver-related mortality. Beyond direct liver-related mechanisms, however, increasing evidence links chronic inflammatory states with all-cause mortality.(16-19) Thus, it is conceivable that SVR eliminates chronic inflammation from HCV which then substantially contributes to reduced all-cause mortality through mechanisms that are still being defined. From a patient and family perspective, all-cause mortality may be a more meaningful outcome of concern rather than disease-specific mortality.
It is evident from this evaluation that a history of hepatic decompensation and albumin levels strongly influence mortality. In patients without a history of decompensation, SVR was associated with an 80% reduction in risk of death compared to a 67% reduction in risk of death for patients with a history of decompensation. In addition, decreasing albumin independently predicted increased mortality regardless of if the patient did or did not achieve SVR. Both findings may serve as additional support for treating patients earlier in their HCV infection, prior to a decompensation or prior to substantial impairment in liver synthetic function. As previously shown in peginterferon-based studies, we confirmed that genotype 3 negatively impacted mortality and may provide additional impetus to treat genotype 3 patients.(1,2) Our findings also serve as a reminder that patients with HCV may have other comorbidities associated with increased mortality, such as congestive heart failure, that may be amenable to intervention regardless of HCV treatment status.
In contrast to studies with peginterferon-based treatment where the risk of HCC was reduced among patients with SVR, existing reports on the risk of HCC occurrence or recurrence after DAA treatment are discordant.(20-27) Some data suggest a higher incidence of HCC occurrence or recurrence and a more aggressive course following successful DAA therapy.(20-23) Others have observed similar HCC rates in treated and untreated patients.(24-27) The disparate data is theorized to reflect differences in patient populations treated with peginterferon therapy versus DAA therapy, the latter being older and having more advanced liver disease
which may inherently increase risk of HCC. Many of these studies, however, are limited by the fact that they did not concomitantly report on the occurrence or recurrence in patients who did not achieve SVR. In this study, patients with SVR after treatment with DAAs had a significantly reduced rate of HCC occurrence compared to patients without SVR after DAAs. The rate of incident HCC observed post-DAA treatment in those achieving SVR was low at 1.9 cases/100 patient years and represented an 84% reduction in HCC cases compared to patients with no SVR. Some reports suggest a time association between DAA initiation and HCC development such that HCC may occur sooner in those treated with DAAs.(20,23) Our data indicated that the time to development of HCC was delayed in those who achieved SVR compared to those without SVR. Moreover, the risk of all-cause mortality was reduced in patients with a history of HCC who achieved SVR after DAA treatment. While the reduction in death rate was less pronounced in those with a history of HCC within 5 years of DAA initiation compared to those with no HCC history, achievement of SVR was still associated with a 64% reduction in death rate compared to those patients with no SVR, highly suggesting that successful treatment with DAAs in the setting of prior HCC was beneficial. This magnitude of reduction was similar to that observed in other smaller trials.(24,25,27)
This study has several limitations. Because the determination of SVR requires laboratory testing 12 weeks or more after EOT, which is an inherent survival advantage for SVR patients, we excluded patients who died before 12 weeks of follow up which may underestimate the death rate for both the No SVR and SVR groups. Because many of our measures of comorbidities rely on ICD-9/10 coding and required a code within 1 year of DAA treatment, it is likely that some diagnoses of interest are underreported, although it is reasonable to assume that the rate of underreporting is similar between SVR and No SVR patients. As previously noted, we did not have complete information about reported cause of death so we cannot address what proportion of the reduction in mortality was related to reduction in liver-related mortality. We did not assess the impact of SVR on HCC recurrence, only incident HCC and death in patients with a recent history of HCC. In patients that developed HCC after DAA treatment we were unable to assess the stage, size or clinical presentation of HCC at diagnosis due to limitations of the administrative database. Given the relatively recent widespread use of interferon-free DAA regimens, our study is limited by the short available follow-up time. The observed differences in risk of death and risk of HCC may change over time as more time from treatment elapses. It is possible that providers more aggressively screen for HCC in No SVR patients compared to SVR patients and consequently more HCC may be diagnosed in No SVR patients. Nonetheless, it is reassuring that all-cause mortality was markedly reduced with SVR such that even if HCC was present but had not been diagnosed, fewer SVR patients apparently died from it. Finally, despite controlling for numerous factors, it is possible that No SVR and SVR patients differ on unmeasured, and possibly nonmeasurable, factors that might account for the observed differences in mortality. However, it is difficult to envision that residual confounding could account for the quite substantial difference in all-cause mortality we observed associated with SVR.
In conclusion, DAA-induced SVR was associated with reduced all-cause mortality among patients with advanced liver disease and HCV genotypes 1, 2, and 3 treated in routine medical practice. These findings strongly support a clinically significant benefit of DAA treatment in patients with advanced liver disease irrespective of HCV genotype. As more people are treated for HCV infection with DAAs, fewer deaths should result in this population.
| |
| |
| |
|
|
|
