A team of clinician-scientists and researchers has developed a new tool that accurately predicts which liver cancer patients are likely to experience a recurrence after surgery.
The machine-learning tool combines genomic and clinical information and outperformed the commonly used TNM staging system, which is based on tumour burden alone, across independent patient groups.
Researchers from the National Cancer Centre Singapore (NCCS), Duke-NUS Medical School (Duke-NUS) and the ASTAR Genome Institute of Singapore (ASTAR GIS) also uncovered two biologically distinct ways in which hepatocellular carcinoma (HCC), the most common form of primary liver cancer, can recur.
The findings could help tailor follow-up care, identify patients most likely to benefit from additional treatment after surgery and design more targeted clinical trials.
The study was conducted through the National Medical Research Council-funded PLANet (Precision Medicine in Liver Cancer across an Asia-Pacific Network) research programme.
Even in patients with early-stage disease, recurrence rates of HCC remain high after surgical resection. Intrahepatic recurrence, where cancer returns within the liver, accounts for 70-80% of all recurrences, while the remaining cases involve distant metastasis, where the cancer spreads beyond the liver to other parts of the body.
Better tools to identify patients at highest risk of recurrence could enable more precise therapies to prevent recurrence after surgery and help match patients to clinical trials most likely to benefit them.
Yet, clinicians currently lack reliable biomarkers to identify which patients are most at risk of recurrence. This makes it difficult to tailor follow-up care, select patients for additional treatment, and design clinical trials efficiently.
To better understand why HCC recurs after surgery, the research team analysed clinical data and conducted comprehensive matched genetic analyses of tumour samples from patients in the PLANet cohort, including recurrent tumour samples from a subset of patients. Of the 106 patients studied, 68 (64.2%) experienced recurrence: 48 in the liver, 11 in other parts of the body, and 9 in both locations.
Two main patterns of recurrence were uncovered:
Polyclonal seeding, in which multiple groups of cancer cells spread simultaneously from the original tumour. This was more often associated with recurrence within the liver. Tumours in this group have distinct characteristics that suggest they may respond better to certain immunotherapies.
Monoclonal seeding occurs when recurrence develops from a single clone from the original tumour. This pattern was more commonly associated with later recurrence and cancer spreading beyond the liver to other parts of the body.
Building on these findings, the team developed the new, machine-learning-based multi-omics tool that has proved to be a much better predictor than the commonly used TNM staging for resected liver cancer, which does not account for cancer genetics, achieving a performance score of 86% compared with 56-68% for TNM.
