Clinical guidelines for screening of liver cirrhosis complications remain, and clinical judgment must prevail.

Without effective treatment against hepatitis C virus (HCV), 60% to 90% of persons with HCV-related cirrhosis will die due to liver complications[1,2]. Studies have shown that successful HCV treatment with direct-acting antivirals (DAA) in patients with cirrhosis leads to a decline in liver-related complications and mortality[3-6]. Yet, recurring questions about people with cirrhosis after successful DAA treatment include identifying those at risk of subsequent liver decompensation and what information can guide medical providers on managing and preventing those complications effectively.

In this issue of AIDS, Corma-Gomez and colleagues present data characterizing the kinetics of the appearance of first liver complications in persons with HCV with advanced liver fibrosis after the achievement of sustained viral response (SVR)[7]. Using a prospective cohort study design including 18 hospitals in Spain, 1,305 persons with advanced liver fibrosis were treated with DAA and followed for a median of 3.6 years from the documented SVR. Approximately 627 had cirrhosis at baseline, of whom 63 had a prior episode of liver decompensation, and 664 were people living with HIV (PLWH). The authors described hepatocellular carcinoma (HCC) and portal-hypertensive complications. Among the latter were ascites, hepatic encephalopathy, portal gastrointestinal bleeding, and hepatorenal syndrome. Fifty-one patients (4.9%) developed a new episode of liver decompensation after a median time of 1.4 years from documented SVR, including 13 with a history of decompensated cirrhosis. The authors defined two kinetic profiles of liver complications, ‘early’ and ‘continuous,’ after a median of 1.1 and 2.1 years from SVR. The ‘early’ presentations included ascites and hepatic encephalopathy, and the ‘continuous’ complications encompassed HCC and portal gastrointestinal bleeding. No cases of ascites or hepatic encephalopathy occurred after three years from SVR. The authors reason that these criteria could guide therapeutic measures for these complications and help identify patients at a higher risk of developing liver-related events so as to continue performing screening and monitoring to prevent liver-related complications.

Corma-Gomez and colleagues’ study has important implications. It highlights that the most important predictor for having a new liver decompensation after SVR is having already had one before DAA treatment[8]. Approximately 1 in 5 individuals with prior liver decompensation developed a new event. Additionally, as in the pre-HCV treatment era, the most frequent manifestations of first episode liver decompensation were ascites, followed by hepatic encephalopathy and infrequently portal hypertensive bleeding. Among PLWH, neither the type of viral hepatitis coinfection nor antiretroviral therapy affects this initial sequence of decompensation[9]. This study provides evidence that the sequence of liver-related complications following SVR with DAA therapy remains unchanged, albeit they occurred much less frequently.

In principle, the characterization of the predictors, sequence, and timing of post-SVR liver complications in patients with cirrhosis could help guide how to target and perhaps scale down screening intensity and treatment during the post-SVR state. However, some considerations may help place the current study in perspective.

First, grouping events in patients with or without liver decompensation adds significant uncertainty to the observed results. Patients with prior decompensation are often monitored more frequently and attentively. Additionally, ongoing treatment of liver complications after SVR in those with previous liver decompensation may modify the clinical presentation of disease events in the post-SVR state[10]. Also, grouping HCC and portal hypertensive complications is problematic as their pathogenesis differs[11,12]. Epigenetic changes associated with HCC development take a long time, but their detection is influenced by screening uptake[13]. This, in part, might explain that 16 of 19 HCC cases were diagnosed in patients without prior know liver decompensation. And 56% of HCC among those without previous liver decompensation occurred within 2-years from SVR.

Second, people with HCV cirrhosis constitute a heterogeneous group even without decompensation. They have different liver physiologic and microstructural impairments, partially captured by the clinical scores used[14]. They also differ in the prevalence of competing risks associated with liver disease progressions, such as hazardous alcohol use or metabolic-associated fatty liver disease, both prevalent among PLWH[15-16]. Cluster survival analysis could mitigate the underlying heterogeneity of the population at risk by identifying optimal clustering of multiple decompensation events that may occur at different time points[17]. Furthermore, when someone post-SVR presents with one outcome, e.g., esophageal hemorrhage, they may be discovered to have undiagnosed encephalopathy, ascites, or HCC before the bleeding. So there is secondary ascertainment of antecedent conditions. This introduces temporal ambiguity into the “time to first event” analysis, and no competing risk analysis was performed in this study. Third, a similar trend of two profiles of temporary kinetics to developed liver decomposition were noted among PLWH, but, in the supplementary figures, after 3 years, curves of liver kinetics overlapped. Key predictors of liver fibrosis progression in PLWH are having CD4 cells less than 200 cells/mm³ and persistent uncontrolled HIV infection[18,19]. All PWLH were on antiretroviral therapy in this cohort, and only 8% had a detectable viral load. No information is provided about CD4 cell count.

The advent of highly effective DAA has spurred optimism for creating new ways to simplify risk stratification of patients with cirrhosis after SVR. Most consider including the use of transient elastography with or without serologic markers [18,20]. Yet, most patients with cirrhosis globally, especially PLWH, are managed by their primary care providers with often limited access to transient elastography[21]. Noteworthy, signs of liver decompensation can take a long time to develop; this can limit interpretation of results if studies do not follow patients at risk for 5-10 years after SVR[22].

Corma-Gomez efforts remind us of the need for large-scale collaborations to achieve a more precision medicine approach. We need more collaborations with longer follow-up, granular descriptions of risk and prognostic factors that affect liver disease progression, and cost-effectiveness analyses of alternative screening and monitoring strategies[23]. Until then, the presented data supports current professional guidelines for HCC screening and periodic endoscopic screening for esophageal varices. Primary care providers should be familiar with and implement current guidelines [24-27]. Management of patients with decompensated cirrhosis cannot be based on a pre-defined time but rely on a careful clinical and individualized management where clinical judgment must prevail.