AI in neuroendocrine liver metastases
Research type
Research Study
Full title
Artificial intelligence in neuroendocrine liver metastases
IRAS ID
363854
Contact name
Andrea Frilling
Contact email
Sponsor organisation
Imperial College London
Duration of Study in the UK
2 years, 11 months, 31 days
Research summary
Neuroendocrine neoplasms are uncommon but challenging tumours that often spread (metastasise) to the liver. Many are metastatic at the point of diagnosis and whilst there is a wide range of possible therapies there is scarce evidence on the optimal ways to treat individual people with neuroendocrine liver metastases (NELM). A significant unmet need in NELM is the lack of factors that can predict long-term survival outcomes, disease recurrence and ways to identify early whether or not patients have responded to treatment. Neither standard blood test-based markers or standard imaging techniques have been demonstrated to be helpful in these ways - the main predictive information available is from a cellular marker called Ki67, which depends on having an invasive procedure such as surgery or biopsy, and its precision may be affected by heterogeneity within the tumour.
Artificial intelligence (AI) approaches applied to medical imaging present interesting new avenues for robust analyses of features in medical scan images and could identify signatures that enhance non-invasive diagnostics, prognostication and treatment response assessment.
This is a pilot study which seeks to evaluate the potential of using AI techniques on MRI scans from patients with NELM (integrated with routine clinical data) to predict long-term outcomes, early recurrence after surgery or response to therapies. The results from this study are expected to provide a valuable foundation for multi-centre cancer research and contribute to the field of imaging-based AI biomarkers in precision oncology.REC name
London - City & East Research Ethics Committee
REC reference
26/PR/0427
Date of REC Opinion
6 May 2026
REC opinion
Further Information Favourable Opinion