Pancreatic cancer is one of the most lethal malignancies worldwide, with a 5-year survival rate of approximately 9%. However, despite significant advances in diagnostic and therapeutic modalities, the prognosis of pancreatic cancer remains dismal. Curative resection is the primary treatment for localised pancreatic cancer; however, the disease often recurs even after successful surgery. Therefore, identifying accurate prognostic factors is crucial for improving patient outcomes and optimising therapeutic strategies. This article aims to provide an overview of the prognostic analysis of curatively resected pancreatic cancer using harmonised positron emission tomography (PET) radiomic features.
Emerging Radiomics: Prognostic Power in Medical Imaging Modalities
Radiomics is an emerging medical imaging field involving the extraction of quantitative imaging features for diagnostic and prognostic purposes. Radiomic features are extracted from various imaging modalities, including computed tomography (CT) and magnetic resonance imaging (MRI). In addition, PET is a nuclear medicine imaging technique that utilises radiolabelled tracers, such as 18F-fluorodeoxyglucose (FDG), to provide functional and metabolic information about tissues. PET has been used to assess various malignancies, including pancreatic cancer, and has shown promise in providing valuable prognostic information.
Harmonisation of Radiomic Features
The lack of standardisation in radiomic feature extraction methods has led to inconsistencies in the reported prognostic value of these features across different studies. Therefore, the harmonisation of radiomic features is essential to improve the reproducibility and reliability of radiomic analysis. Harmonisation involves the implementation of standardised protocols for image acquisition, reconstruction, preprocessing, and feature extraction. In PET imaging, harmonisation efforts include standardised uptake value (SUV) normalisation, the implementation of standardised segmentation methods, and the adoption of standard radiomic feature extraction software packages.
Prognostic Analysis of Curatively Resected Pancreatic Cancer
Several studies have investigated the prognostic value of PET radiomic features in curatively resected pancreatic cancer. These studies have reported that specific radiomic features are significantly associated with overall survival (OS), disease-free survival (DFS), and recurrence-free survival (RFS).
- As measured by PET radiomic features, tumour heterogeneity has been associated with a poor prognosis in pancreatic cancer. In addition, high intratumoral heterogeneity, as indicated by increased entropy, irregular shapes, and non-uniform intensity distributions, has been linked to aggressive tumour biology and higher recurrence rates.
- Elevated metabolic activity, as indicated by high maximum and mean standardised uptake values (SUVmax and SUVmean), has been correlated with poor survival outcomes in pancreatic cancer. This is likely because high metabolic activity reflects aggressive tumour behaviour and the presence of viable tumour cells after resection.
- PET-derived textural features, such as coarseness, contrast, and homogeneity, have been shown to hold prognostic value in pancreatic cancer. In addition, these features provide information about the spatial distribution of tracer uptake within the tumour, which can indicate tumour aggressiveness and potential for recurrence.
- By incorporating PET radiomic features with clinicopathological factors, researchers have developed prognostic models to predict survival and recurrence in curatively resected pancreatic cancer patients. These models have shown improved predictive accuracy compared to models based on clinicopathological factors alone.
Conclusion
The integration of harmonised PET radiomic features in the prognostic analysis of curatively resected pancreatic cancer has shown promising results. By providing valuable information on tumour biology and aggressiveness, PET radiomics may aid in tailoring personalised treatment plans and improving patient outcomes. Future research should focus on validating these findings in larger, multicenter studies and developing standardised radiomic feature extraction methods to enhance these results’ reproducibility and clinical applicability. Additionally, the integration of radiomic features from other imaging modalities, such as CT and MRI, may provide a more comprehensive understanding of tumour characteristics and further enhance prognostic accuracy. Ultimately, the incorporation of harmonised PET radiomic features into clinical practice has the potential to significantly impact the management of pancreatic cancer patients and contribute to the advancement of precision oncology.
You are here: home »