# The Role of Computed Tomography-Determined Total Tumor Volume at Baseline in Predicting Outcomes of Patients with Locally Advanced Unresectable or Metastatic Pancreatic Ductal Adenocarcinoma

## TL;DR

This study shows that measuring total tumor volume via CT scans helps predict survival in advanced pancreatic cancer patients, especially when combined with other biomarkers.

## Contribution

The study introduces a survival prediction model combining computed tumor volume with biological markers for advanced pancreatic cancer.

## Key findings

- Patients with total tumor volume over 400 cm³ had significantly shorter overall survival (9.4 vs. 13.0 months).
- A combined model of tumor volume and biomarkers achieved a c-index of 0.64 for overall survival prediction.
- High-risk patients based on the model had worse progression-free and overall survival outcomes.

## Abstract

Pancreatic adenocarcinoma remains one of the most aggressive cancers, and predicting prognosis and treatment response is a challenge. This study explores the utility of measuring the total tumor volume of primary and metastatic lesions on routine imaging scans for estimating survival in patients with advanced pancreatic cancer. We first analyzed patients who received standard chemotherapy and compared survival based on tumor volume and then developed a score that combines tumor volume with other biomarkers to predict survival. We found that patients with very large total tumor volumes have lower survival rates. When tumor volume was combined with tumor and inflammatory markers, the model provided a clearer way to distinguish patients at higher or lower risk of disease progression of death. These findings suggest that total tumor volume, together with biological markers, may guide treatment decisions in advanced pancreatic cancer.

Background: Total tumor volume (TTV), derived from imaging data, has emerged as a potential prognostic biomarker in various cancers. This study aimed to evaluate the impact of TTV on outcomes in advanced pancreatic ductal adenocarcinoma (PDAC) and to validate a survival prediction model combining TTV with baseline clinico-biological markers. Materials and Methods: We conducted a retrospective analysis of 150 patients with locally advanced or metastatic PDAC treated with first-line FOLFIRINOX from 2010 to 2021. TTV was calculated by manually segmenting all visible lesions on baseline CT scans. Progression-free survival (PFS) and overall survival (OS) were the primary endpoints. A cut-off value for TTV predicting 6-month PFS was determined in 140 patients using AUC and Youden’s Index and then applied to OS analysis. A multivariate Cox regression model incorporating TTV, CA 19-9, and neutrophil-to-lymphocyte ratio (NLR) was developed in 94 patients to establish a survival risk score. Results: 12,028 lesions were annotated. OS was slightly but significantly different between TTV above and below the median value of 69.60 cm3 (12.4 vs. 13.5 months, p = 0.0269). A cut-off of 400 cm3 distinguished two groups: patients with TTV > 400 cm3 had significantly shorter OS (9.4 months) compared to those with TTV ≤ 400 cm3 (13.0 months, p = 0.0056). A similar trend was observed for PFS, though not statistically significant (7.4 months for TTV > 400 cm3 vs. 8.2 months for TTV ≤ 400 cm3, p = 0.0735). The combined model achieved a mean c-index of 0.62 for PFS and 0.64 for OS. Based on the risk score, high-risk patients had significantly worse median PFS (5.5 vs. 9.2 months, p = 0.0008) and median OS (7.2 vs. 13.5 months, p < 0.0001). Conclusions: TTV is a valuable prognostic marker in advanced PDAC. A model integrating TTV with biological markers enhances survival prediction and supports risk stratification in clinical practice.