This is a single-centre, prospective cohort study. Consecutive patients with compensated liver disease and hepatocellular carcinoma (HCC) for curative-intent treatment will be invited to this study. The study follow-up duration will be five years. The primary outcome is the composite of incident high-risk varices, hepatic decompensation and liver-related mortality. The participants will undergo transient elastography at baseline and then half-yearly post-HCC curative treatment; and oesophagogastroduodenoscopy (OGD) at baseline and then at year 1, 3 and 5 post-HCC curative treatment. Clinical assessments and laboratory tests will also be done at baseline and every 6 months until year 5 to identify any clinical events.

1.1 Significance of chronic liver disease and hepatocellular carcinoma

Hepatocellular carcinoma (HCC) burdens the global healthcare system with a drastic increase in incidence by 70% from 1990 to approximately 747,000 cases and 480,000 deaths in 2019.

Despite implementation of screening ultrasound in at-risk populations and advances in the treatment of HCC in recent decades, HCC remains a lethal malignancy with a median survival of less than a year. Apart from the oncological factors, the underlying liver disease has an important impact on morbidity and mortality in patients with HCC as most HCC develop under the setting of chronic hepatic inflammation and fibrosis from patients with advanced chronic liver diseases of various aetiologies.

1.2 Non-invasive measurement of portal hypertension

In the Baveno VI consensus, the term "compensated advanced chronic liver disease" (cACLD) described spectrum of chronic liver diseases ranging from advanced liver fibrosis to compensated cirrhosis. Patients with cACLD are at risk of hepatic decompensation and development of HCC, especially for those with clinically significant portal hypertension (CSPH), which is conventionally measured invasively by hepatic venous pressure gradient (HVPG). The recently updated Baveno VII consensus implemented a non-invasive approach by incorporating liver stiffness measurement (LSM) by vibration-controlled transient elastography (VCTE) and platelet count in ruling in or out CSPH in patients with cACLD. In the Baveno VII consensus and as well incorporated into the latest practice guidance on portal hypertension, LSM ≤ 15kPa and platelet count ≥ 150 x 109/L rules out CSPH, whereas LSM > 25kPa, LSM 20-25kPa with platelet count <150 x 109/L or LSM 15-20kPa with platelet count <110 x 109/L rule in CSPH, prompting the presence of gastroesophageal varices. As well in Baveno VI consensus, LSM < 20 kPa with platelet count > 150 x 109/L also implies a minimal risk of high-risk varices and thus can have a screening oesophagogastroduodenoscopy (OGD) spared.

The development of spleen stiffness measurement (SSM) by VCTE also revolutionised the field by demonstrating its correlation with portal pressure and CSPH. Studies have shown that, by combining Baveno VII criteria with SSM, the accuracy in identifying CSPH (such as the presence of high-risk varices) was enhanced compared to adopting Baveno VII criteria alone.

1.3 Unmet need 1: Conflicting data on LSM/SSM in prediction of hepatic events in HCC patients

Use of LSM and SSM in prediction of HCC recurrence was reported in a few previous studies. The studies involved a one-stop LSM and/or SSM measurements with longitudinal follow-up on the recurrence rate of HCC. For instance, high baseline LSM was shown to be a poor prognostic factor in patients with HCC receiving radiofrequency ablation, and achieving low LSM reduced HCC recurrence. Another prospective study involving 175 patients with HCC suitable for resection showed that SSM was the only predictor for late HCC recurrence.

However, there are lack of data in the prediction of hepatic decompensation in patients with HCC. First, studies aiming to investigation on risk prediction of decompensation either excluded patients with HCC or regarded HCC as one of the study endpoints. Second, the present data available also showed conflicting results. For instance, two retrospective studies performed from our group suggested that Baveno VII criteria (i.e. LSM by VCTE and platelet count) could accurately identify high risk varices and hepatic events in both advanced HCC and that of different Barcelona Clinic Liver Cancer (BCLC) stages. On the contrary, another retrospective suggested that the use of both Baveno VI or VII criteria were inaccurate in predicting the presence of high-risk varices or CSPH measured by HVPG across different stages of HCC. Heterogeneity was present in the study subjects and data analysis among different studies, leading to divergent results. SSM as well was not included in these studies and that the effect of combined Baveno VII and SSM in prediction of decompensation in HCC patients is still uncertain.

Inclusion Criteria:

• Aged 18 or above
• Known chronic liver disease(s)
• HCC for curative-intent treatment (defined by HCC diagnosed with typical radiological features or histology, and planned for surgical resection or local ablative therapy as a curative-intent treatment)

Exclusion Criteria:

• Current or history of decompensated liver cirrhosis (i.e. Child's C cirrhosis, prior decompensating events such as ascites, variceal bleeding, hepatic encephalopathy and hepatorenal syndrome)

  o Child's B cirrhosis without decompensating events is not excluded

• Past history of HCC (ie. The current HCC is a recurrence of priorly treated HCC or a second de novo HCC after previous first HCC)

• Non-primary liver tumour (such as secondary liver tumour due to metastasis from another distant primary tumour)

• History of liver transplantation or plan for liver transplantation as the modality of curative-intent HCC treatment

• Asplenism or history of splenectomy

• Contraindication to OGD (eg. Intestinal perforation of obstruction)

• Serious medical illness with limited life expectancy of less than 6 months

• Pregnancy

• Unable to obtain or refusal of informed consent from patient

Consecutive adult patients with hepatocellular carcinoma fulfilling the inclusion and exclusion criteria will be invited to join the study.

Data may be shared upon reasonable request.

6 months after the first publication until 15 years after the end of the study

By email communication