The study was led by investigators from the National Cancer Centre Singapore (NCCS), Duke-NUS Medical School and the ASTAR Genome Institute of Singapore (ASTAR GIS) through the National Medical Research Council-funded PLANet programme, or Precision Medicine in Liver Cancer across an Asia-Pacific Network. The findings, published in Gut on 21 July 2026, also reveal that HCC can return through two biologically distinct routes. These insights could help clinicians identify patients who need closer surveillance, select individuals for additional treatment after surgery and design more focused clinical trials.

HCC is the most common form of primary liver cancer and remains a major cause of cancer mortality worldwide. It is the third leading cause of cancer-related death globally and disproportionately affects Asian populations, where approximately 72 per cent of cases occur. In Singapore, liver cancer is the third most common cause of cancer death among men and the fifth among women. The region’s disease burden is strongly linked to chronic hepatitis B virus infection, which can drive long-term inflammation, liver damage and malignant transformation.

Surgical removal of the tumour can offer patients a chance of long-term survival, particularly when the disease is detected at an earlier stage. However, recurrence remains common even after apparently successful resection. Around 70 to 80 per cent of recurrences occur within the liver, while the remainder involve metastasis to distant organs. Current clinical staging systems can estimate risk based on tumour size, number, vascular invasion and spread, but they provide limited information about the genetic diversity and evolutionary behaviour of individual cancers.

To investigate why some tumours return, the researchers analysed clinical information and performed comprehensive matched genetic studies on tumour samples collected through the PLANet cohort. The analysis included samples from recurrent tumours in a subset of patients, enabling the team to compare the original cancer with the disease that later re-emerged. Among 106 patients included in the study, 68, or 64.2 per cent, developed recurrence. Forty-eight patients experienced recurrence inside the liver, 11 developed disease elsewhere in the body and nine had both intrahepatic and distant recurrence.

The genetic comparisons identified two principal patterns of tumour spread. In the first, known as polyclonal seeding, several genetically distinct groups of cancer cells appear to leave the original tumour and establish new growths at the same time. This mechanism was more frequently associated with recurrence within the liver. The findings suggest that these tumours contain multiple malignant subpopulations, some of which may interact differently with the immune system and could be more responsive to selected immunotherapies.

The second pattern, monoclonal seeding, occurs when recurrence is driven primarily by a single cancer-cell clone from the original tumour. This route was more commonly linked to later recurrence and the spread of cancer beyond the liver. A surviving clone may possess biological properties that allow it to remain dormant, resist treatment or adapt to a new tissue environment before forming a detectable metastasis. Distinguishing between these mechanisms may therefore provide information about both the timing and likely location of recurrence.

Building on the genetic findings, the investigators created a multi-omics prediction model that integrates information from several biological layers. The tool considers clinical features such as tumour size and cancer stage, blood markers and genomic measurements, together with the activity of a 15-gene signature strongly associated with recurrence. Machine-learning methods allow these variables to be combined into a single risk estimate, potentially revealing patterns that would be difficult to recognise from any individual measurement.

The model was evaluated in three independent cohorts, including data from The Cancer Genome Atlas Liver Hepatocellular Carcinoma dataset. This validation was important because the TCGA group is largely Western, while the PLANet programme focuses on Asia-Pacific populations. Across the evaluations, the tool achieved an area under the receiver operating characteristic curve of approximately 86 per cent, compared with reported values of 56 to 68 per cent for TNM staging. A higher area under the curve indicates that a model is better able to distinguish patients at higher risk from those at lower risk, although further prospective testing will be needed before routine clinical use.

The researchers say the tool could eventually support more personalised surveillance and treatment after surgery. Patients predicted to be at high risk might be prioritised for intensive monitoring or clinical trials of adjuvant systemic therapies, while those at lower risk could avoid unnecessary treatment and its associated toxicities. The team is also using spatial sequencing to examine the tumour microenvironment and identify potential drug targets. Additional data, including CT imaging, may be incorporated into future versions of the model. Together, these efforts could improve understanding of how HCC evolves and help match patients with therapies designed for the specific biology of their disease.

Subject of Research: Human tissue samples

Article Title: Clonal diversity underpins distinct modes of recurrence in hepatocellular carcinoma: the PLANet cohort study

News Publication Date: 3 August 2026

Web References: https://doi.org/10.1136/gutjnl-2026-338227; https://www.nccs.com.sg; https://www.a-star.edu.sg/gis; https://www.duke-nus.edu.sg/

References: National Registry of Diseases Office. (2026). Singapore Cancer Registry Annual Report 2023. Ministry of Health Singapore. Chan S, Sun H, Xu Y et al. “The Lancet Commission on addressing the global hepatocellular carcinoma burden: comprehensive strategies from prevention to treatment.” The Lancet. 2025;406:731–778.

Keywords: hepatocellular carcinoma, liver cancer, cancer recurrence, machine learning, multi-omics, genomics, precision medicine, tumour evolution, polyclonal seeding, monoclonal seeding, hepatitis B virus, PLANet cohort

Hepatocellular carcinoma remains a formidable clinical challenge globally due to its high recurrence rates even after curative treatments such as surgical resection. Although transarterial chemoembolization serves as an adjuvant therapy to improve outcomes postoperatively, recurrence prediction remains imprecise, complicating follow-up and treatment planning. Against this backdrop, the development of a reliable predictive tool using noninvasive imaging biomarkers offers a potential paradigm shift in managing postoperative HCC.

At the heart of this research lies the application of radiomics, an emerging field that extracts a vast array of quantitative features from medical images, revealing subtle information beyond what is visually perceptible. This study capitalized on multiparametric MRI, acquiring multiple sequences that capture diverse biological characteristics of tumor tissue. By integrating these complex imaging features using advanced computational strategies, the investigators sought to distill an imaging signature predictive of tumor recurrence.

The research was conducted retrospectively across two medical institutions involving 149 patients with confirmed HCC treated by PA-TACE. This cohort was methodically divided into training, internal validation, and external validation groups, ensuring the robustness and generalizability of the results. The use of multiple centers not only adds credibility but also reflects the model’s potential applicability across diverse clinical settings.

Radiomics features were meticulously extracted from three distinct MRI sequences. Each sequence provides unique tissue contrasts, capturing information on tumor heterogeneity and microenvironment alterations linked with aggressive disease behavior. The high-dimensional data extracted required sophisticated selection methods to prevent overfitting and identify the most prognostically relevant features, a task accomplished using the Least Absolute Shrinkage and Selection Operator Cox regression (LASSO-COX).

The LASSO-COX technique, known for its ability to handle multicollinearity and high-dimensional predictors, narrowed down an expansive feature set to 15 optimal radiomic variables. These features, when combined into a singular Rad-score, represented a quantitative imaging biomarker predictive of recurrence risk. The median Rad-score established a threshold distinguishing patients with divergent prognoses following PA-TACE treatment.

Beyond imaging, clinical parameters played a pivotal role in refining the prediction model. Among the numerous clinical variables analyzed, the neutrophil-to-lymphocyte ratio (NLR) and tumor size emerged as independent predictors significantly associated with recurrence-free survival. The NLR, a marker of systemic inflammation, has gained attention for its prognostic relevance across malignancies, reflecting the complex interplay between tumor biology and host immune response.

Tumor size, a long-recognized prognostic factor, reflects tumor burden and likelihood of occult metastases or vascular invasion, both of which portend a higher recurrence probability. Integrating these clinical parameters with the Rad-score yielded a composite nomogram, a user-friendly tool that transforms complex predictive analytics into individualized risk assessments.

Performance metrics underscored the model’s predictive strength. The concordance indices (C-index) measuring agreement between predicted and actual outcomes consistently exceeded 0.82 across training and validation cohorts, highlighting the model’s discriminative power. Receiver operating characteristic (ROC) curves elucidated the time-dependent accuracy of the model, substantiating its potential utility in clinical decision-making.

Calibration curves further demonstrated excellent concordance between predicted probabilities and observed recurrence rates, an essential attribute for any predictive model aspiring to clinical integration. This alignment suggests that clinicians can rely on the nomogram’s outputs to enhance postoperative surveillance strategies and potentially guide adjuvant treatment decisions tailored to individual patient risk profiles.

The implications of this study extend beyond simple prognostication. By leveraging multiparametric MRI—a noninvasive, widely available imaging modality—alongside computational radiomics, this model offers a precision medicine approach that can dynamically capture tumor biology. Such insight can facilitate early intervention at recurrence and potentially improve long-term survival outcomes for HCC patients.

Furthermore, the study exemplifies how interdisciplinary collaboration between radiology, oncology, computational science, and clinical epidemiology converges to address complex oncological challenges. It underscores the importance of integrating diverse data streams to forge predictive tools that transcend traditional staging systems, which often fail to encapsulate tumor heterogeneity and host factors adequately.

While the study’s retrospective design introduces inherent limitations, the inclusion of an external validation cohort strengthens confidence in the model’s applicability across institutions. Future prospective studies and clinical trials will be imperative to confirm these findings and evaluate how this radiomics-based nomogram can be incorporated into routine clinical workflows.

In the era of artificial intelligence and big data, the successful application of radiomics for personalized oncologic prognosis heralds a new chapter for imaging biomarkers. This research not only underscores the transformative potential of multiparametric MRI but also paves the way for radiomics-driven tools to become integral components of cancer management paradigms worldwide.

Ultimately, these advancements are poised to optimize postoperative management of hepatocellular carcinoma by enabling clinicians to identify high-risk patients who might benefit from intensified surveillance or adjunctive therapies. Such precision-guided care promises to improve patient quality of life and eventual survival through timely and targeted interventions.

As the oncology community continues to embrace innovative methodologies, studies like this provide invaluable blueprints demonstrating how quantitative imaging data can revolutionize traditional clinical prognostication. This holistic approach represents a critical step forward in realizing personalized medicine for liver cancer patients.

The confluence of multiparametric imaging and radiomics with robust clinical variables transforms the landscape of postoperative HCC care. With ongoing refinement and validation, such models hold immense promise for broader oncology applications, ultimately leading to more informed, data-driven therapeutic decisions.

Researchers and clinicians eagerly await further developments and validation studies to unlock the full potential of radiomics-based nomograms in hepatocellular carcinoma and beyond. This landmark study exemplifies the burgeoning synergy between technology and medicine aiming to improve outcomes in one of the world’s most challenging malignancies.

Subject of Research: Prediction of hepatocellular carcinoma recurrence after postoperative adjuvant transarterial chemoembolization using multiparametric MRI-based radiomics combined with clinical data.

Article Title: Multiparametric MRI-based radiomics and clinical nomogram predicts the recurrence of hepatocellular carcinoma after postoperative adjuvant transarterial chemoembolization.

Article References:
Guo, X., Song, J., Zhu, L. et al. Multiparametric MRI-based radiomics and clinical nomogram predicts the recurrence of hepatocellular carcinoma after postoperative adjuvant transarterial chemoembolization. BMC Cancer 25, 683 (2025). https://doi.org/10.1186/s12885-025-14079-y

Image Credits: Scienmag.com

DOI: https://doi.org/10.1186/s12885-025-14079-y

HCC represents the third leading cause of cancer-related mortality across the globe, with alarmingly high postoperative recurrence rates that can soar up to 70%. The complexity of accurately predicting recurrence in HCC is largely attributable to the tumor’s intricate microenvironment, particularly the spatial heterogeneity observed within the tumor immune microenvironment (TIME). This dynamic interplay among tumor cells, immune cells, and various components of TIME presents significant obstacles in risk assessment and prognostication.

In the course of their investigation, the researchers successfully developed what they termed the tumor immune microenvironment spatial (TIMES) score. This quantitative scoring system offers a detailed characterization of the spatial distribution of immune cell populations within the tumor microenvironment. The methodology underlying the TIMES score involved the employ of the XGBoost machine learning algorithm, which was trained on an extensive multiplex immunofluorescence dataset derived from samples provided by 61 HCC patients. This approach exemplifies a synergistic fusion of computational biology and clinical research, leveraging advanced analytics to enhance predictive capabilities.

The TIMES system distinguishes itself by providing a holistic assessment of tumor-immune interactions, calling upon the integration of whole-slide imaging (WSI) with an AI-driven spatial analysis algorithm. This fusion of technologies empowers the system to deliver precise recurrence risk predictions grounded in the spatial expression profiles of five pivotal biomarkers: SPON2, ZFP36L2, ZFP36, VIM, and HLA-DRB1. This multi-dimensional analysis has unveiled significant insights into the immune landscape of HCC, paving the way for advancements in patient stratification and personalized medicine.

A particularly noteworthy finding from the study was the identification of SPON2 as the most predictive biomarker. Its expression pattern, particularly within natural killer (NK) cell subsets, exhibited a robust correlation with HCC outcomes. The spatial immune profiling demonstrated a pronounced disparity between non-recurrent and recurrent HCC patients, with the former group showing a notable enrichment of CD57+ NK cells positioned at the invasive tumor margin. Such regional immune heterogeneity is crucial as it provides prognostic information that traditional histopathological grading systems may overlook.

To further illuminate the biological underpinnings of their findings, the researchers delved into the molecular mechanisms by which SPON2 modulates NK cell function. Utilizing three-dimensional migration assays, they established that SPON2 significantly fosters the directional migration of NK cells towards tumor cells. Complementary cytotoxicity assays revealed that SPON2+ NK cells displayed markedly enhanced cytolytic activity, which was correlated with a substantial increase in the activation levels of CD8+ T lymphocytes. Notably, experiments in NK cell-specific SPON2-knockout mouse models revealed diminished interferon-gamma (IFN-γ) secretion and compromised NK cell infiltration, both of which contributed to expedited tumor growth. These results affirm that SPON2+ NK cells belong to a highly active subset that plays a pivotal role in curbing HCC recurrence.

The predictive power of the TIMES system was validated within an independent cohort, where it achieved an impressive accuracy of 82.2% and a specificity of 85.7%. Such performance metrics not only underscore the efficacy of the TIMES system but also highlight its superiority in comparison to existing clinical prediction models that currently guide therapeutic decision-making in HCC settings. This level of precision is a game-changer in the oncological landscape, granting clinicians newfound confidence in tailoring treatment strategies based on individualized recurrence risks.

In a bid to enhance clinical utilization, the research team has established an open-access online tool that empowers clinicians to input standard immunohistochemistry-stained images and obtain comprehensive reports that detail TIMES scores alongside personalized assessments of recurrence risk. This resource has the potential to streamline patient management protocols and enable oncologists to make informed decisions around postoperative monitoring and treatment interventions.

Importantly, the algorithms and computational frameworks that underpin the TIMES system have been patented, signaling a transition from laboratory research to translational application. The researchers are actively seeking partnerships within the industry to standardize protocols that can facilitate the swift adaptation of the TIMES system within clinical practices, thereby emphasizing the importance of bridging the gap between scientific discovery and real-world application.

Overall, this research not only provides a tangible predictive tool that can enhance clinical decision-making but also enriches our understanding of the immune mechanisms that underpin HCC recurrence. As the field of oncology continues to evolve, the implications of the findings presented by Prof. SUN Cheng and his team may lay the groundwork for novel immunotherapeutic strategies targeting SPON2+ NK cells, heralding a new era of specialized interventions aimed specifically at improving prognosis for HCC patients.

This breakthrough in the understanding of HCC’s complex immune interactions illustrates not just a remarkable achievement in cancer research but also a promising horizon for developing innovative cancer therapies tailored to individual patient profiles. By refining predictive models and enhancing risk stratification, this avenue of research embodies the confluence of artificial intelligence and immunology, offering fresh pathways toward unprecedented improvements in cancer care and patient outcomes.

Ultimately, the revelations provided by this study signify a substantial leap forward in our ability to confront HCC recurrence and may serve as a beacon of hope amid the ongoing battle against cancer. With the establishment of the TIMES system, the future of personalized medicine in hepatocellular carcinoma appears promising, bearing the potential to redefine standard practices in oncology and significantly improve the quality of life for patients facing this formidable disease.

Subject of Research: Prediction of hepatocellular carcinoma recurrence
Article Title: Spatial immune scoring system predicts hepatocellular carcinoma recurrence
News Publication Date: 12-Mar-2025
Web References: Nature
References: None available
Image Credits: None available

Keywords

Hepatocellular carcinoma, tumor immune microenvironment, predictive modeling, SPON2, natural killer cells, immunotherapy, machine learning.