Safety and efficacy of directly‐acting antiviral therapy for chronic hepatitis C virus in elderly people

Abstract Introduction In Italy, the prevalence of hepatitis C virus (HCV) infection is higher in the elderly, although the efficacy and safety of treatment in this population has not been extensively studied. Moreover, little is known about how much pharmacological interaction affects eligibility to treatment and to what extent the treatment affects subsequent outcomes. Methods We retrospectively analyzed the efficacy and safety of directly acting antivirals (DAAs), drug‐to‐drug interactions, and post‐treatment outcomes in 138 patients with HCV aged 70 years or older, who were consecutively treated in our center between 2015 and 2020. Results The mean age was 77 years old (range = 70–95 years old). The Cumulative Illness Rating Scale of pretherapy severity was classified as moderate to severe in 65% of patients. Fifty‐five patients (40%) presented compensated cirrhosis, eight of which were complicated by hepatocellular carcinoma (HCC) and all were cured before treatment. One hundred two patients (74%) were taking two or more drugs (range = 0–5 concomitant drugs registered) and in 29 patients (21%) we found potential drug‐to‐drug interaction. In 11 of those 29 patients (38%), we were forced to change the chronic therapy, when all therapeutic regimens were equal in terms of efficacy and interactions, to avoid potentially serious drug interactions. One serious adverse event occurred in our sample population (i.e., diverticular bleeding due to interaction with direct oral anticoagulants [DOACs]), whereas mild side effects occurred in 37% of patients. The undetectability of HCV RNA at the end of treatment was achieved in 97% of patients, whereas a sustained virological response (SVR) 12 and SVR 24 were obtained in 98% of patients. When comparing pretherapy with post‐therapy data, after a medium follow‐up of 15 months (median = 1 year, minimum = 2 months, and maximum = 4 years), we observed a reduction in the incidence of episodes of liver decompensation in patients with cirrhosis and a slight increase in the incidence of HCC (with 6 recurrent and 5 de novo HCC), diagnosed within 13 months from the end of therapy. In all patients, we found a significant improvement in all ultrasound variables and a significant reduction in the elastographic measurements. No significant differences in outcomes were observed dividing the population into patients aged ≥ 80 and < 80 years old. Conclusions Directly acting antiviral therapy was found to be safe and effective in elderly people, and, despite the large number of concomitant drugs, pharmacological interactions appeared to not affect the adherence to therapy or the incidence of adverse events. Side effects were mostly independent from the type of DAA used and from the burden of comorbidity. In long‐term follow‐up, the benefit of DAA therapy mainly concerned liver pathology and should be strongly advised in patients with cirrhosis. The therapy was found to not affect extrahepatic comorbidities but allowed to end follow‐up in noncirrhotic patients with savings in terms of resources. Finally, patients should not be excluded based on age if they have a good performance status.


| INTRODUC TI ON
The mean age of patients with chronic hepatitis C virus (HCV) infection and the number of elderly patients with advanced liver disease is gradually increasing in many countries, including Italy. [1][2][3][4][5][6] Many screening strategies have been implemented, although they always concern high-risk groups, excluding the elderly population from the analysis. This risk strategy is unsuccessful in the perspec- Many studies have also shown that the age in itself and the age of onset of HCV infection significantly impact the degree of fibrosis: older individuals have an increased risk of progressing to cirrhosis and of developing complications. [7][8][9][10][11][12][13][14] Older patients have always been considered difficult to treat and not suitable for interferon therapies. [15][16][17][18] However, the second generation of directly acting antiviral drugs (DAAs) and the expansion of the prescribing criteria, have made all patients with a known infection virtually suitable for antiviral treatments. Some studies have shown that DAAs are highly effective, safe, and tolerable even in elderly patients. Most of them are based on extrapolated data from large cohorts of phase III studies and the few real-life studies have been limited to analyzing the efficacy and safety without providing follow-up data. [19][20][21][22][23][24][25][26][27][28][29] Many clinicians do not begin DAA treatment in the elderly, as they are concerned of the many concomitant drugs and their possible interactions. Furthermore, it is still unclear which is the impact that antiviral therapy has on extrahepatic outcomes and the longterm benefit of viral eradication in this population needs further studies. Few cost-utility studies have been specifically applied to treating hepatitis C in the elderly, 30 and, because of comorbidities and a shorter life expectancy, the cost-effectiveness ratio increases with age making the therapy less attractive for elderly patients.
For all these reasons, these patients have often been neglected by avoiding treatment, although the current guidelines do not consider age as an element of choice in starting treatment. 31,32 The elderly, in the perspective of a global eradication of the disease, should be considered as a real special population to screen and treat like other subjects considered at high risk. Searching for more evidence in this area is fundamental for the correct allocation of health resources and health equality.
This study aims to evaluate not only the efficacy and safety of DAA treatment on a real-life population of patients with HCV but also to assess the impact of drug-to-drug interactions on the eligibility for treatment and on the natural history of liver disease. Furthermore, we analyzed the trend of hepatic and extrahepatic variables following treatment to define the overall outcome of elderly patients.

| MATERIAL S AND ME THODS
In this retrospective study, we selected patients aged 70 years or older at the time of the start of treatment from a cohort of 575 outpatients treated with new-generation direct antivirals from 2015 to 2020 at the Division of Gastroenterology, Policlinico Umberto I, Rome (Italy).

| Inclusion criteria
• Patients ≥ 70 years of age with chronic HCV infection eligible for DAA therapy. Eligibility for treatment was assessed according to current clinical practices, assessing the general health conditions, potential drug interactions, and the patient's willingness to undergo treatment. Only an 85-year-old patient in poor general condition and poor prognosis was excluded from access to treatment. differences in outcomes were observed dividing the population into patients aged ≥ 80 and < 80 years old.
Conclusions: Directly acting antiviral therapy was found to be safe and effective in elderly people, and, despite the large number of concomitant drugs, pharmacological interactions appeared to not affect the adherence to therapy or the incidence of adverse events. Side effects were mostly independent from the type of DAA used and from the burden of comorbidity. In long-term follow-up, the benefit of DAA therapy mainly concerned liver pathology and should be strongly advised in patients with cirrhosis. The therapy was found to not affect extrahepatic comorbidities but allowed to end follow-up in noncirrhotic patients with savings in terms of resources. Finally, patients should not be excluded based on age if they have a good performance status.

K E Y W O R D S
DAAs, efficacy, elderly, HCV, safety

| Exclusion criteria
• Patients < 70 years old at the beginning of treatment.
From an overall cohort of 575 patients with chronic HCV infection, we selected 138 patients who were over 70 years old at the start of treatment (24%). We collected the following data before and after DAA therapy: • Personal and anamnestic data, in particular, age, gender, body mass index (BMI), comorbidities, concomitant drug therapy, and Cumulative Illness Rating Scale (CIRS). 33 • Parameters related to liver disease, such as the presence of cirrhosis, duration of illness, and episodes of decompensation.
• Ultrasound and laboratory parameters.
• Parameters related to DAA therapy, such as duration, drug interactions 34 and side effects.
• Cause of death, if deceased during therapy or follow-up.
The pretherapy blood biochemical data was compared with the data collected 6 months after the end of therapy. If the data at 6 months was not available, we considered the data collected 3 months after the end of therapy.
Each patient underwent an ultrasound scan before and after therapy. The pre-therapy ultrasound data was compared with the data of the last ultrasound in our possession after the end of the therapy (if comparable quantitative measures were reported).
Elastometry with Acoustic Radiation Force Impulse (ARFI) or Fibroscan was evaluated before and after therapy, comparing ARFI with ARFI and Fibroscan with Fibroscan. The medium time from the end of therapy and the ultrasound/elastometry was 409 days. Comparable ultrasound scans were available for 111 patients (81%) and elastometry for 105 patients (76%, in 66 evaluated with ARFI and in 39 with Fibroscan).
The clinical conditions, concomitant drugs, and comorbidities were extrapolated from the last visit before starting therapy and after the end of antiviral therapy, covering a medium follow-up of 15 months (median = 1 year, minimum = 2 months, and maximum = 4 years). From these data, we retrospectively calculated the CIRS score and data on mortality.

| Statistical analysis
The data collected were analyzed with NCSS2020 software.
Quantitative variables were described in terms of mean, standard deviation, and range, whereas qualitative variables were expressed as frequencies and percentages. Comparisons of numerical variables were performed with the t test for paired data, whereas qualitative variables with the chi-square test. A logistic regression analysis was performed to determine if there were any variables associated with the onset of side effects. A P value below 0.05 was considered significant.

| Baseline characteristics of patients
The baseline characteristics of the patients of our sample population are summarized in Table 1 When present, the main concomitant causes of liver damage were alcohol (10%) and steatosis (9.4%). The most represented genotypes of our dataset were 1B (56.5%) and 2 (36.2%). Ninety-eight patients (71%) were naïve to any treatment, whereas 40 patients had already been exposed to previous therapies, of which 38 were to an interferon-based therapy and two to a DAA treatment (who were both relapsers).
Fifty-five patients (40%) were cirrhotic and in almost all cases they had preserved liver function (Child-Pugh A 94%; medium model of end stage liver disease [MELD] score = 9, range = 6-20; refer to Table 1 for further details). The prevalence of episodes of pretherapy liver decompensation was 16% (1 bleeding, 7 ascites, and 1 encephalopathy). Nine patients were diagnosed with HCC before treatment Thirty-seven patients (26.8%) had mixed cryoglobulinemia. Of these, only seven were symptomatic for neuropathy and purpura.

| Safety and efficacy of treatment
The treatment schedules as well as the data relative to the safety and efficacy of treatment are summarized in Table 2. Potential interactions between the antiviral therapy and patients' drug therapy were observed in 29 patients (21%) but only in 11 subjects was it necessary to change the home therapy to avoid potentially serious interactions. The most used DAA regimes were Ledipasvir/Sofosbuvir (17%), Glecaprevir/Pibrentasvir (24%), and Grazoprevir/Elbasvir (25%). In terms of length of treatment, most patients (65%) followed a 12-week treatment, followed by an 8-week treatment (21%). A minority of patients followed a 16-week treatment (3%) or a 24-week therapy (11%).
In two patients, we observed an adverse event (AE) due to drugto-drug interaction (1.4%). In both cases, it was caused by an interaction with direct oral anticoagulants (DOACs). One patient developed spontaneous skin ecchymosis without other serious consequences.
The other one experienced diverticular bleeding with hospitalization and consequent early stop of the antiviral therapy after 1 month from the start. The patient died 2 years later of heart failure without ever repeating HCV RNA.
There were no other serious adverse events (SAEs). One minor side effect occurred in 37% of patients, of which the most frequently reported were nausea/loss of appetite (11%) and asthenia (16%).
When we correlated each single side effect with each single type of DAA regime in univariate analysis (chi-square test), we found sig-  Table 3 for details). The baseline presence of cirrhosis or elevated CIRS appeared to be protective for the onset of nausea/loss of appetite ( Table 3).
The efficacy of therapy, evaluated both with the intention-totreat and for per-protocol analysis, was excellent with an end of treatment (EOT) response in 98% of patients and a rate of sustained virological response (SVR) 12 and SVR 24 of 98% (refer to Figure 3 for details).

| Post-therapy outcome
A total of 12 patients (8.7%) died during follow-up, nine of which were cirrhotic (  Table 4).
In patients with cirrhosis, the incidence of post-treatment de-  Table 4 for details).
We compared the blood chemistry tests before and after treatment and found a statistically significant improvement in the value of albumin, total and direct bilirubin, ALT, and AST (Table 5).
A slight reduction in alpha-fetoprotein levels was observed in 77% of patients with a medium decrease of −0.18 × upper limit of normal (ULN).
It is worth underlining that the majority of patients presented normal alpha-fetoprotein levels before and after therapy.  (Table 6).
Thirteen hepatic nodules were found following treatment (   Table 8 for more details).

| DISCUSS ION
From our analysis, it emerges that treatment with DAAs is safe and effective even in elderly patients. Although only 6% of patients did not have comorbidities and only 9% did not take any type of concomitant drug therapy, polycomorbidities, and drug polytherapy were not found to be obstacles to start antiviral treatment. The incidence of side effects also had little effect on the beginning of antiviral treatment. Although largely present (37% of patients experienced at least one), the side effects were mild AEs in all cases (but two), which did not require any type of medical management or suspension of treatment.
The efficacy in terms of SVR was thus equal to that obtained in younger patients. However, our data was affected by the number of

TA B L E 7 Hepatocellular carcinoma
The hepatic outcomes following therapy showed an improvement both in terms of ultrasound findings and in terms of blood tests (refer to Tables 5 and 6). In addition, we observed a very low incidence of decompensation events in post-therapy patients with cirrhosis (6.25%), especially considering that the risk of decompensation for patients with over 20 years of illness is normally in the order of 66% every year. Another factor to consider is that these decompensation events in our series were almost entirely attributable to patients with hepatocarcinoma.
The risk of post-therapy HCC was 8%, which was higher compared to pretherapy, but was still globally low compared to the quota (22%) reported in previous studies concerning the prevalence of HCC in elderly patients without SVR. 7-10 Likewise, all-cause mortality was in the order of 8% in our dataset compared to 26%, as previously reported, at 10 years without SVR. [7][8][9][10] When considering extrahepatic outcomes, we found an improvement in the CIRS score, which was mainly driven by noncirrhotic patients (F0-F2 s Metavir) whose comorbidity score increased due to chronic liver disease, a factor that disappeared following treatment with the consequent decrease in the score itself. The aforementioned patients represented 40% of our cohort and were all able to end follow-up, due to the absence of other hepatic comorbidities.
The same results were confirmed even when a subanalysis was conducted dividing the patients by age (patients ≥ 80 years old vs < 80 years old). No significant differences in the basal characteristics of comorbidities and pharmacotherapy, in the incidence of side effects, and in the response to treatment were observed (Table 8). Furthermore, there was no higher incidence of decompensation events in patients aged over 80 years old compared with younger patients following treatment, although the overall mortality was higher (but not significantly higher) in this cohort of patients.

| CON CLUS IONS
The benefit of DAA therapy in elderly patients was found to mainly concern liver disease and is strongly indicated in patients with cirrhosis, regardless of age. Concerning noncirrhotic patients, the therapy did not appear to affect extrahepatic comorbidities but Areas of interest to be explored remain the evaluation of the impact of antiviral therapy on the quality of life and the execution of a cost-utility analysis, which would be crucial for the correct allocation of health resources.

ACK N OWLED G EM ENTS
This study was supported by research funding of La Sapienza, University of Rome (Italy) and submitted as ePoster at EASL ILC2021.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to declare.