Predicting Survival After Surgical Resection for the Entire Spectrum of Anatomically Resectable HCC

Completed

• Conditions: Hepatocellular Carcinoma Resectable

• Registration Number: NCT06294808
• Lead Sponsor: National Cancer Centre, Singapore

Brief Summary

Clinical outcomes after surgical resection in HCC is a continuum and is clearly related to tumor burden but needs better definition. The researchers describe the use of the "metro ticket" approach to analyze surgical outcomes over the whole spectrum of anatomically resectable HCC to define overall survival including intermediate stage tumors (BCLC B). The analysis the researchers provide in this study enables the clinician to select the optimal surgical resection candidate based on robust long term survival data.

In addition, study compares outcome for open surgery vs laparoscopic surgery, survival outcome for viral and non-viral HCC using Albumin-Bilirubin (ALBI) for more comprehensive study result.

Detailed Description

Patients who underwent surgical resection (both open surgery and laparoscopic surgery) for HCC from 1st January 2000 to 30th June 2018 by the joint hepato-pancreato-biliary surgery service at the Singapore Healthcare Group of Hospitals (Singapore General Hospital and the National Cancer Centre Singapore) were retrospectively identified from a prospectively kept institutional database.

The assessment of the size, number of nodules and vascular invasion of HCC were based on pre-operative CT/MRI imaging and pathological assessment of the resected specimens. All resections were histologically confirmed as HCC. Patients with macrovascular invasion and extrahepatic invasion were excluded from analysis as were patients resected for other palliative intents (e.g. ruptured HCC).

Study Timeline

• Study Start: 2017-07-17
• Primary Completion: 2020-06-30
• Study Completion: 2020-06-30
• Study First Submitted: 2024-02-26
• Status Verified: 2024-03
• Study First Submitted QC: 2024-03-04
• Last Update Submitted: 2024-03-04
• Study First Posted: 2024-03-06
• Last Update Posted: 2024-03-06

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 1043

Inclusion Criteria

1. Unequivocal diagnosis of Hepatocellular Carcinoma (HCC) by histology
2. Patients who underwent surgical resection for HCC

Exclusion Criteria

1. Patients who underwent liver resection for other malignancies ie cholangiocarcinoma
2. Patients with HCC who did not undergo liver resection.
3. Patients with HCC who underwent liver transplant.

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
5-year mortality.	At least 5 years.	Time between the respective surgery to the time of death. Bivariate contour plots of 5-year survival probability as a function of X = AFP concentration (ng/mL) and Y = tumor burden index (size of largest nodule + 2\*number of nodules) (TBI) were obtained as smoothed contours originating from an empirically-derived bivariate 5-year survival distribution S(X,Y)=1- F(x ≤X,y ≤Y). The function F(x ≤X,y ≤Y) is the empirical, bivariate cumulative distribution of 5-year mortality as a function of X and Y defined over the domain (0\<X≤5000,1≤Y≤15) in our patient cohort. All patients in our study cohort were followed up for at least 5 years following respective surgery which allowed estimation of actual all-cause 5-year mortality.
Concordance of pre-operative radiological and post operative surgical specimen.	Through study completion, an average of 3 years.	The metro ticket modeling was based on tumour size and number of nodules extracted at pre-operative radiological imaging and post surgical pathology assessment. Continuous variables are summarized as mean, standard deviation, 25th and 75th percentiles, median and minimum and maximum. Categorical variables are summarized as counts (N) and percentages (%).
Agreement between TBI calculated from pre-operative imaging versus those obtained from post-surgical pathological assessment.	Through study completion, an average of 3 years.	Bland-Altman analysis, which is based on a simple but effective graphical approach for evaluating bias and for calculating bounds on the expected magnitude of the individual differences arising between the two measurement methods, is used. A 45° line of identity with scatter plot was produced to display the correlation between the 2 measurements. Bias was evaluated by calculating (d) ̅ the mean difference of the measurements. Limits of agreement, calculated as (d) ̅ ±2sd , where sd is the standard deviation of the individual differences, defined bounds within which 95% of the measurement differences are expected to fall. The B-A plot defines the limits of agreements. B-A analysis is performed using TBI over a wide range, 0 ≤ TBI ≤ 30, and a restricted range focusing on small tumors, 0 \< TBI ≤ 6.

Trial Locations

Locations (2)

National Cancer Centre, Singapore; 🇸🇬 Singapore, Singapore

Singapore General Hospital; 🇸🇬 Singapore, Singapore