172P - Survival prediction using CT based radiomic features in patients of pancreatic cancer treated by chemotherapy followed by SBRT: A pilot study

Background

Pancreatic cancer is a disease with poor prognosis. Stereotactic body radiation therapy (SBRT) is emerging hope to achieve better resectability and local control. Pancreatic cancer treatment results are currently poorly predicted by the best clinical prediction algorithms. Through quantitative imaging analysis, radiomics and dynamic imaging characteristics can offer data on clinical outcomes and build clinical models based on imaging phenotypes or radiomics signatures. The objective of this pilot study was to analyze the association of CT based radiomic features with overall survival (OS) in pancreatic cancer patients treated with SBRT.

Methods

Ten patients of borderline resectable and locally advanced pancreatic cancer were included. All patients received neoadjuvant chemotherapy followed by SBRT and further chemotherapy and then assessed for surgery. Using pyRadiomics, radiomic features of gross tumour volume were extracted using contrast enhanced planning CT images acquired during SBRT planning. All the statistical analysis was performed using R software. Survival rates were analyzed using Kaplan-Meier survival curves.

Results

Three of the 10 patients were resected. Median follow-up was 15 months (range,5-24 months), median OS was 25 months. A total of 851 radiomic features [shape, first order, gray-level co-occurrence matrix (GLCM); grey level size zone matrix (GLSZM), gray-level distance zone matrix (GLDZM), grey level run length matrix (GLRLM) and grey level zone length matrix (GLZLM), gray-level co-occurrence matrix (GLCM); the neighborhood gray-level different matrix (NGLDM)] were extracted. Based on the radiomics score, patients were stratified into low and high risk categories.

Conclusions

This pilot study highlights the potential of CT-based radiomic features in predicting survival in patients with pancreatic cancer treated with chemotherapy followed by SBRT. The identified radiomic features, along with clinical parameters, offer valuable prognostic information that can aid in treatment decision-making and patient counseling. However, a larger number of patients are required to validate the results.

Editorial acknowledgement

Clinical trial identification

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.