Spatial Conformation of Pancreatic Cancer Is Associated with Disease Recurrence after Curative-Intent Total...

Area of Science:

• Oncology
• Computational Pathology
• Cancer Research

Background:

• Pancreatic ductal adenocarcinoma (PDAC) often recurs after neoadjuvant therapy and surgery.
• Current histopathologic assessments of treatment response in PDAC are limited, especially for minor response cases.
• Understanding residual tumor biology is crucial for improving patient outcomes.

Purpose of the Study:

• To investigate if the spatial organization of residual PDAC after total neoadjuvant therapy (TNT) provides prognostic information.
• To develop spatial risk models for predicting disease-free survival (DFS) in PDAC patients with minor pathologic response.

Main Methods:

• Retrospective analysis of 203 resected PDAC patients with minor pathologic response.
• Whole-slide H&E images were segmented into cancer and stroma.
• Quantification of spatial composition and configuration of residual tumor architecture.

Main Results:

• Treatment resistance was linked to a fragmented, interface-rich tumor ecology with higher edge density and reduced aggregation.
• Two multivariable spatial risk models, based on cancer/stroma shape and edge density, independently predicted DFS.
• Spatial models outperformed traditional treatment response assessments in predicting outcomes.

Conclusions:

• Residual cancer-stroma topology offers independent prognostic signals in post-TNT PDAC.
• Quantifying spatial architecture can identify patients at higher risk of recurrence.
• These findings support the development of spatially informed adjuvant strategies and mechanistic studies.