Angiogenesis regulators S100A4, SPARC and SPP1 correlate with macrophage infiltration and are prognostic biomarkers in colon and rectal cancers.

Introduction: Increasing evidence suggests that it is necessary to find effective and robust clinically validated prognostic biomarkers that can identify "high-risk" colorectal cancer (CRC) patients. Currently, available prognostic factors largely include clinical-pathological parameters and focus on the cancer stage at the time of diagnosis. Among cells of tumor microenvironment (TME) only Immunoscore classifier based on T lymphocytes showed high predictive value. Methods: In the present study, we performed the complex analysis of mRNA and protein expression of crucial regulators of tumor angiogenesis and tumor progression, expressed by tumor-associated macrophages (TAMs): S100A4, SPP1 and SPARC. Colon and rectal cancer patients were investigated independently and in a combined cohort (CRC). For mRNA expression, we analyzed RNA sequencing data obtained from TCGA (N=417) and GEO (N=92) cohorts of colorectal cancer patients. For protein expression, we performed IHC digital quantification of tumor tissues obtained from 197 patients with CRC treated in the Department of abdominal oncology in Clinics of Tomsk NRMC. Results: High S100A4 mRNA expression accurately predicted poor survival for patients with CRC independently of cancer type. SPARC mRNA level was independent prognostic factors for survival in colon but not in rectal cancer. SPP1 mRNA level had significant predictive value for survival in both rectal and colon cancers. Analysis of human CRC tissues revealed that S100A4, SPP1 and SPARC are expressed by stromal compartments, in particular by TAMs, and have a strong correlation with macrophage infiltration. Finally, our results indicate that chemotherapy-based treatment can change the predictive direction of S100A4 for rectal cancer patients. We found that S100A4 stromal levels were higher in patients with better response to neoadjuvant chemotherapy/chemoradiotherapy, and S100A4 mRNA levels predicted better DFS among non-responders. Discussion: These findings can help improve the prognosis of patients with CRC based on S100A4, SPP1 and SPARC expression levels.