RIKEN's Groundbreaking Advance in Liver Cancer Risk Prediction

Researchers at Japan's premier research institute, RIKEN, have unveiled a transformative tool in the fight against hepatocellular carcinoma (HCC), the most lethal form of liver cancer. Led by Senior Research Scientist Xian-Yang Qin at the RIKEN Center for Integrative Medical Sciences (IMS), the team developed a machine-learned biomarker known as the MYCN niche score. This innovative score identifies individuals at high risk for developing liver tumors even in tumor-free liver tissue, offering hope for earlier intervention and prevention.

Hepatocellular carcinoma arises primarily from chronically damaged livers, often due to viral hepatitis, alcohol abuse, or increasingly, metabolic dysfunction-associated steatohepatitis (MASH, formerly known as NASH). Globally, HCC claims over 800,000 lives annually, with recurrence rates post-treatment reaching 70-80 percent, largely because of late detection. In Japan, where liver cancer has historically been a leading cause of cancer death due to high hepatitis C virus (HCV) prevalence, incidence stands at around 30,000 new cases yearly, though declining thanks to antiviral therapies. However, the rise in MASH-related cases poses new challenges, with prevalence of MASH cirrhosis climbing to 14.6 percent in recent surveys.

This RIKEN study, published in the Proceedings of the National Academy of Sciences (PNAS) on February 18, 2026, integrates spatial transcriptomics—a technique mapping gene expression in tissue sections with precise location data—with machine learning algorithms to pinpoint precancerous microenvironments.

Understanding the MYCN Oncogene's Role in Liver Tumorigenesis

MYCN, or MYC N-myc proto-oncogene, is a transcription factor gene typically associated with neuroblastoma but increasingly implicated in solid tumors like HCC. In liver cancer, MYCN expression marks cancer stem-like cells (CSCs), which drive tumor initiation, metastasis, and recurrence. Prior studies from Qin's lab showed MYCN overexpression in HCC tissues correlates with poor prognosis, and acyclic retinoid (ACR) therapy selectively targets MYCN-positive CSCs to prevent recurrence.

The new research demonstrates MYCN's direct oncogenic function. Using a hydrodynamic tail vein injection transposon system, researchers overexpressed MYCN in mouse livers. Alone, it caused no tumors; combined with constitutively active AKT—a kinase promoting cell survival—72 percent of mice developed HCC-like tumors within just 50 days. These tumors mirrored human HCC histologically and molecularly, validating the model.

This synergy highlights how MYCN exploits damaged liver environments, such as those in MASH, to fuel tumorigenesis. AKT activation simulates chronic liver injury signals, creating a fertile 'niche' for MYCN-driven cancer.

RIKEN IMS: Hub for Cutting-Edge Cellular Research

RIKEN, Japan's flagship research organization, fosters collaborations with universities like Jikei University School of Medicine and Gifu University, where Qin has affiliations. The IMS Laboratory for Cellular Function Conversion Technology focuses on reprogramming diseased cells for therapy, aligning perfectly with biomarker discovery for regenerative medicine.

Such interdisciplinary work positions RIKEN at the forefront of precision oncology. For aspiring researchers in Japan, opportunities abound in research jobs at institutes like RIKEN, bridging academia and clinical application.

Spatial Transcriptomics Meets Machine Learning: The Methodological Innovation

The study's power lies in spatial transcriptomics, which profiles thousands of genes while preserving spatial context, unlike bulk RNA sequencing. In a MASH mouse model, researchers tracked MYCN dynamics over time, revealing elevated MYCN in tumor-free regions forming a distinct 'niche'—a cluster of 167 differentially expressed genes fostering tumor-permissive conditions.

Machine learning then analyzed this data. Training on spatial patterns, the algorithm scored gene expression profiles with 93 percent accuracy for MYCN niche presence. This 'niche score' quantifies risk without needing tumor tissue.

• Spatial mapping identifies precancerous hotspots.
• ML model processes complex multi-omics data for predictive scoring.
• Step-by-step: Tissue sectioning → gene profiling → niche clustering → score computation.

Defining the MYCN Niche: A Precancerous Signature

The MYCN niche comprises genes upregulated in MYCN-high regions, promoting stemness, inflammation, and fibrosis. This microenvironment precedes visible tumors, explaining high recurrence. In mice, niche formation tracked liver damage progression, peaking before tumorigenesis.

Visualized via immunofluorescence, tumors glow with GFP-MYCN (green), myr-AKT (red), cytokeratin (white), underscoring co-localization.

Check the full RIKEN press release for figures.

Human Validation: Prognostic Power in HCC Patients

Applying the score to public human HCC datasets (e.g., TCGA), higher MYCN niche scores strongly predicted recurrence and survival, especially from non-tumor liver biopsies. Non-tumor scores outperformed tumor ones, ideal for post-surgical risk stratification.

In Japan, where 5-year HCC survival hovers around 40 percent despite advances, this could personalize surveillance.

Clinical Implications: From Prediction to Prevention

The MYCN niche score enables risk stratification for high-risk patients, like those with MASH or post-HCV cure. Pairing with serum MYCN ELISA—validated earlier by the team—offers non-invasive monitoring. ACR, effective against MYCN+ CSCs, could target high-score patients preventively.

Broader impacts: Shift from reactive treatment to proactive prevention, reducing Japan's liver cancer burden amid rising metabolic diseases. Explore career advice for medical researchers advancing such biomarkers.

Building on RIKEN's Legacy in Liver Cancer Research

Qin's team pioneered serum MYCN as a recurrence biomarker in 2024, linking it to ACR response in phase III trials. High-throughput screening identified MYCN inhibitors, paving therapeutic paths. Collaborations with Japanese universities amplify impact.

Future Outlook: Dissecting Mechanisms and Therapies

Qin envisions decoding niche biology for targeted drugs: "We aim to dissect mechanisms of cancer-permissive environments." Clinical trials for niche score-guided therapy loom, potentially revolutionizing HCC management in Japan and beyond.

Read the PNAS paper: Oncogenic function and transcriptional dynamics of MYCN.

Photo by Markus Winkler on Unsplash

Career Opportunities in Japan's Biomedical Research

This breakthrough underscores demand for experts in transcriptomics and AI. Platforms like higher-ed research jobs, university jobs, and professor positions in Japan offer paths to contribute. Explore Rate My Professor for insights into leading labs.