Advanced liver disease in Russian children and adolescents with chronic hepatitis C

Summary Russia has one of the highest prevalences of paediatric chronic hepatitis C infection (CHC). Our aim was to provide a detailed characterization of children and adolescents with CHC including treatment outcomes. Thus, an observational study of children with CHC aged <18 years was conducted in three hepatology centres from November 2014 to May 2017. Of 301 children (52% male), 196 (65%) acquired HCV vertically, 70 (23%) had a history of blood transfusion or invasive procedures, 1 injecting drug use and 34 (11%) had no known risk factors. Median age at HCV diagnosis was 3.1 [interquartile range, IQR 1.1, 8.2] and 10.8 [7.4, 14.7] at last follow‐up. The most common genotype was 1b (51%), followed by 3 (37%). Over a quarter of patients (84, 28%) had raised liver transaminases. Of 92 with liver biopsy, 38 (41%) had bridging fibrosis (median age 10.4 [7.1, 14.1]). Of 223 evaluated by transient elastography, 67 (30%) had liver stiffness ≥5.0 kPa. For each year, increase in age mean stiffness increased by 0.09 kPa (95% CI 0.05, 0.13, P < 0.001). There was significant correlation between liver stiffness and biopsy results (Tau‐b = 0.29, P = 0.042). Of 205 treated with IFN‐based regimens, 100 (49%) had SVR24. Most children (191, 93%) experienced adverse reactions, leading to treatment discontinuation in 6 (3%). In conclusion, a third of children acquired HCV via nonvertical routes and a substantial proportion of those with liver biopsy had advanced liver disease. Only half of children achieved SVR24 with IFN‐based regimens highlighting the need for more effective and better‐tolerated treatments with direct‐acting antivirals. Further studies are warranted in Russia on causes and prevention of nonvertical transmission of HCV in children.


| INTRODUC TI ON
Russia has one of the highest burdens of HCV infection globally with an estimated seroprevalence of 4.3% and the highest absolute number of chronic infections of 3.2 million. 5 In 2013, a CHC prevalence of 336 per 100 000 population was reported in the HCV national surveillance system, reaching 670 per 100 000 in some regions; overall CHC incidence was 40 per 100 000 per year but varied widely across regions, from 13 to 68. 6 Over the last few years, the reported CHC incidence has gradually decreased to 35-36 per 100 000 in 2016-2017. 7 Genotype (GT) 1b is the most prevalent genotype accounting for 55% of all cases, followed by GT3 (35%). 8 In 2013, the reported CHC prevalence was 14 per 100 000 children aged 1-14, 6 with the corresponding incidence of 1.75 per 100 000 children. 9 Similar to adults, there is a downward trend observed over the last years with the reported incidence of 1. 33-1.44 per 100 000 children in 2016-2017. 7 Vertical transmission was the most common mode of acquisition accounting for two-thirds of new infections, with anti-HCV prevalence among pregnant women reported as 1.3% in 2011-2012. 6 Route of transmission was not determined for 21% of all infected children aged ≤14 years. 6 Our aim was to provide a detailed characterization of children and adolescents with chronic HCV with regard to mode of acquisition, HCV genotype, clinical status, type of treatment and treatment outcomes in three paediatric hepatology centres in Russia in the pre direct-acting antivirals (DAA) era.

| MATERIAL S AND ME THODS
An observational study of children and adolescents with chronic A total of 100 children seen consecutively in each centre and meeting eligibility criteria were included. 10 The eligibility criteria were diagnosis with chronic HCV aged <18 years and being in routine follow-up at the participating centres.
Pseudonymized individual-patient data were retrospectively collected according to a standard protocol using the secure, web-based REDCap electronic data capture tool. 11 Additional data collection on treatment response and adverse events for the sub-group of children still undergoing treatment in July 2015 was conducted prospectively until May 2017, to obtain treatment outcome data.
Ethics approval for the study was obtained from the National Scientific and Practical Center of Children's Health of the Ministry of Health of the Russian Federation, Moscow.

| Definitions
Chronic HCV was defined as detectable HCV RNA in two or more blood samples at least 6 months apart. For children not vertically infected, duration of diagnosed infection was calculated as time from date of diagnosis.
The mode of transmission was reported by participating clinics as vertical (mother is known to have HCV), suspected 205 treated with IFN-based regimens, 100 (49%) had SVR24. Most children (191, 93%) experienced adverse reactions, leading to treatment discontinuation in 6 (3%).
In conclusion, a third of children acquired HCV via nonvertical routes and a substantial proportion of those with liver biopsy had advanced liver disease. Only half of children achieved SVR24 with IFN-based regimens highlighting the need for more effective and better-tolerated treatments with direct-acting antivirals. Further studies are warranted in Russia on causes and prevention of nonvertical transmission of HCV in children.

K E Y W O R D S
adolescent, biopsy, child, chronic hepatitis C, elastography, liver fibrosis, Russia healthcare-associated infection (history of receipt of the blood products or invasive procedures in healthcare settings and no maternal history of HCV), injecting drug use or unknown.
Normal ranges of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) are age-dependent and changed during the study period. We used the cut-off for AST and ALT of 40 IU/L, to allow comparison with the literature. The lower limits of normal for albumin were taken as 35 g/L, platelets 150 × 10 9 /L. 12 Liver biopsy scoring systems Knodell and Metavir were used for liver fibrosis staging, which was categorized as no fibrosis, portal fibrosis (Knodell fibrosis score 1, Metavir F1), bridging or significant fibrosis (Knodell fibrosis score 3, Metavir F2-3) and cirrhosis (Knodell fibrosis score 4, Metavir F4). 13,14 Liver biopsies were reviewed at each centre by experienced histopathologists.
Heights were converted to height-for-age z-scores (HAZ) using the WHO Growth Standard for measurements at age <5 years old, 17 and WHO 2007 growth reference at age 5-18 years. 18

| Statistical analysis
Categorical data were summarized as frequency (percentage) and con-

| Liver biopsy
Ninety-two (31%) children had a liver biopsy (four children had two), with median age at last biopsy of 9.4 [IQR 6.1, 13.0] years ( Table 1) Cirrhosis was identified at liver biopsy in one 15-year-old adolescent in whom infection acquired through contaminated blood products was diagnosed 7 months prior to biopsy. Rates of bridging fibrosis or cirrhosis at the last biopsy are summarized in Table 1.

| Transient elastography
A total of 345 TE measurements were reported in 223 (74%) children. Results of the last TE are summarized in Table 2. There was a significant association between older age at TE and increased liver stiffness (r = 0.22, P = 0.001) ( Figure 1). Moderate increase in liver stiffness (≥7 kPa) was not seen in children <7.5 years of age.
In 84 children with >1 TE, median time between first and last TE was 18.7 [6.4, 19.5] months. In patients with ≥1 TE who completed treatment before the last TE, mean liver stiffness decreased from TA B L E 1 Bridging fibrosis and cirrhosis detected on liver biopsy by patient characteristics 5.2 (SD 1.7) to 4.6 (SD 1.6) between the first and last TE (P < 0.001) (Supporting Information Figure S3).

| Comparison of biopsy and transient elastography results
Sixty-four children had both a biopsy and a scan. For 34, scanning occurred within a year of a biopsy (Figure 2), with a significant but relatively weak correlation between TE and biopsy results (Tau-b = 0.29,
Indications for the last treatment were evidence of fibrosis on liver  (Table 3). Six children required treatment with antidepressants and one required thyroxin replacement. Across all 205 treated children, 54 (26%) had dose reduction and/or discontinuation.

| D ISCUSS I ON
Reasons for premature discontinuation a  IQR, interquartile range. a More than one adverse event or discontinuation reason may be reported in the same child. Multiple adverse events were reported in the six children who discontinued due to drug reactions including hypersensitivity (n = 2), weakness (n = 2), depression/anxiety (n = 4), anaemia (n = 2), alopecia (n = 3) and insomnia (n = 1). present, it is not possible to confidently identify rapid CHC progressors with non-invasive tests, and liver biopsy remains a gold stan- it is associated with substantial toxicity and has suboptimal efficacy.
Over two-thirds of children in this study were treated. Only 57% of children treated with PegIFN/RBV achieved SVR24, similar to the results of a meta-analysis demonstrating 58% of overall SVR across genotypes; rates of treatment discontinuation due to adverse events were also similar. 31 The proportion of children with SVR24 was lowest in Moscow, which might be explained by their more complex patients, and also their treatment of children from across Russia, with joint follow-up with local hepatitis C clinics potentially contributing to more frequent treatment discontinuations. This highlights the difficulties of treating children with prolonged IFN-based schemes, especially those living in remote settings.
Most treated children in our study experienced adverse events.
Of those treated with PegIFN/RBV, one in three developed anaemia or neutropenia, one in three had hair loss and one in six suffered from anxiety/depression; the latter may be underestimated given are in a particularly difficult situation as they are excluded from voluntary licensing agreements and compulsory licensing is challenging to implement. 35,36 We showed that patients completing treatment had a significant decrease of liver stiffness measured by TE, suggesting that treatment in childhood maybe beneficial. In

ACK N OWLED G EM ENTS
We thank children and their families and clinicians looking after them, who made this study possible.