IMMUNOREACT 4: Peritumoral Microenvironment Associated with Anastomotic Leaks After Surgery for Rectal Cancer

Area of Science:

• Oncology
• Immunology
• Surgical Pathology

Background:

• Anastomotic leaks (ALs) are significant complications following rectal cancer surgery.
• Local immune dysregulation in the tumor microenvironment is increasingly implicated in impaired anastomotic healing.
• The study investigates the immune characteristics of histologically normal rectal mucosa adjacent to tumors as a potential factor in AL development.

Purpose of the Study:

• To explore the immune microenvironment of tumor-adjacent rectal mucosa.
• To identify immune markers associated with the risk of postoperative anastomotic leaks (ALs).
• To develop and evaluate predictive models for ALs integrating immune and clinical data.

Main Methods:

• Analysis of a prospective (n=121) and retrospective (n=262) cohort using flow cytometry and immunohistochemistry.
• Comparison of immune markers, including epithelial activation and lymphocyte subsets, between patients with and without ALs.
• Development of exploratory predictive models for ALs using immune and clinical variables.

Main Results:

• Flow cytometry identified moderate predictive potential for ALs using markers like CK+HLAabc+ MFI, CD8+CD38+ cells, and CD3+CTLA4+ cells.
• Immunohistochemistry showed predictive ability for ALs with markers such as CD3+, CD8+, CD8β+, and Tbet+.
• A combined model including CD8β+, BMI, neutrophil/lymphocyte ratio, and tumor location achieved an AUC of 0.67 for AL prediction.

Conclusions:

• Immune activation in histologically normal, tumor-adjacent rectal mucosa, marked by epithelial HLA upregulation and specific T cell infiltration, correlates with postoperative ALs.
• While predictive accuracy is currently limited, these findings suggest a tumor-conditioned immune microenvironment may impair anastomotic healing.
• Integrating mucosal immune profiling with clinical data offers a promising avenue for AL risk assessment, requiring further validation.