Abstract
In PDAC, CA accumulation, which is related to tumor vascularization and perfusion, inversely correlates with tumor cellularity. The standard of care GEM treatment results in decreased CA accumulation, which impedes drug delivery. Further investigation is warranted into potentially detrimental effects of GEM in combinatorial therapy regimens.
Introduction
Pancreatic ductal adenocarcinoma (PDAC) presents a significant clinical challenge due to its complex, "fortress-like" tumor microenvironment. Characterized by poor vascularization and high interstitial fluid pressure, these tumors often resist systemic therapies by physically blocking drug delivery. Currently, clinicians lack established non-invasive imaging biomarkers to stratify patients or monitor physiological changes during treatment. While gemcitabine (GEM) has remained a cornerstone of PDAC therapy for decades, its specific impact on tumor hemodynamics has been debated. This study investigates whether contrast agent (CA) accumulation in CT and MRI can serve as a proxy for tumor cellularity and how standard treatments like GEM alter the tumor's "delivery infrastructure."
Methods
The researchers utilized a multi-modal approach integrating retrospective human data with prospective murine models to validate imaging biomarkers.
Imaging Biomarkers
Two primary metrics were evaluated:
- Hounsfield Units ratio (HUr): Used in CT scans to measure CA "pooling" in the tumor relative to the aorta.
- Area Under the Curve (AUC60r): Used in Dynamic Contrast-Enhanced (DCE) MRI to assess tumor perfusion relative to muscle tissue.
Study Cohorts
The study comprised several human and mouse cohorts (Figure 1):
- Human Cohort h1: 35 therapy-naive patients used to correlate CT/MRI imaging with histological findings.
- Human Cohort h2: 151 patients with locally advanced PDAC monitored during chemotherapy (GEM-based vs. FOLFIRINOX).
- Murine Models (m1-m4): 56 tumors from various Genetically Engineered Mouse Models (GEMM) were used for longitudinal monitoring and drug delivery analysis.
Advanced Analysis
To precisely map drug delivery, the team employed Laser Ablation–Inductively Coupled Plasma–Mass Spectrometry (LA-ICP-MS), using cisplatin as a surrogate tracer for chemotherapy penetration into the tumor tissue.
Results
Imaging as a Proxy for Cellularity
A key finding was the inverse correlation between CA accumulation and tumor cellularity. High CA accumulation (bright scans) indicated stroma-rich, low-cell-density environments, while low accumulation (dark scans) indicated dense, cell-packed tumors. In humans, stroma-rich regions (hPDAClow) showed a mean HUr of 0.64, significantly higher than the 0.34 observed in cell-rich (hPDAChigh) regions (p < 0.001).
The Impact of Gemcitabine
Gemcitabine treatment was found to significantly reduce tumor vascularization. In humans, GEM led to an 18.5% decrease in CA accumulation in initially well-perfused tumors. In murine models, GEM reduced CA accumulation (AUC60r) by 34% and slashed cisplatin delivery by nearly 70% (from 3.1 mg/g to 0.92 mg/g, p = 0.004). Histological analysis confirmed that GEM reduces the proliferation of endothelial cells (22.3% vs. 30.9% in controls), effectively "choking off" the blood supply.
Comparison with FOLFIRINOX
In contrast to GEM, the FOLFIRINOX regimen did not significantly reduce CA accumulation. This suggests that FOLFIRINOX may be less damaging to the tumor's vascular network, potentially preserving the delivery route for combined agents.
Discussion
This study identifies HUr and AUC60r as viable non-invasive biomarkers for PDAC stratification. The discovery that gemcitabine induces hypo-vascularization has profound implications for combination therapies. If GEM restricts the very blood supply needed to deliver subsequent drugs, the timing and sequencing of drug administration in clinical protocols may require urgent re-evaluation.
Limitations include the retrospective nature of the human data and the relatively small MRI subset. Future prospective trials are needed to validate the clinical cutoffs (e.g., HUr > 0.36) established here and to explore optimized scheduling for combinatorial PDAC treatments.
Summary updated