Fueling the flames of colon cancer - does CRP play a direct pro-inflammatory role?

Abstract

Background: Systemic inflammation, diagnostically ascribed by measuring serum levels of the acute phase reactant C-reactive protein (CRP), has consistently been correlated with poor outcomes across cancer types. CRP exists in two structurally and functionally distinct isoforms, circulating pentameric CRP (pCRP) and the highly pro-inflammatory monomeric isoform (mCRP). The aim of this pilot study was to map the pattern of mCRP distribution in a previously immunologically well-defined colon cancer (CC) cohort and explore possible functional roles of mCRP within the tumor microenvironment (TME).

Methods: Formalin-fixed, paraffin-embedded (FFPE) tissue samples from 43 stage II and III CC patients, including 20 patients with serum CRP 0-1 mg/L and 23 patients with serum CRP >30 mg/L were immunohistochemically (IHC) stained with a conformation-specific mCRP antibody and selected immune and stromal markers. A digital analysis algorithm was developed for evaluating mCRP distribution within the primary tumors and adjacent normal colon mucosa.

Results: mCRP was abundantly present within tumors from patients with high serum CRP (>30 mg/L) diagnostically interpreted as being systemically inflamed, whereas patients with CRP 0-1 mg/L exhibited only modest mCRP positivity (median mCRP per area 5.07‰ (95%CI:1.32-6.85) vs. 0.02‰ (95%CI:0.01-0.04), p<0.001). Similarly, tissue-expressed mCRP correlated strongly with circulating pCRP (Spearman correlation 0.81, p<0.001). Importantly, mCRP was detected exclusively within tumors, whereas adjacent normal colon mucosa showed no mCRP expression. Double IHC staining revealed colocalization of mCRP with endothelial cells and neutrophils. Intriguingly, some tumor cells also colocalized with mCRP, suggesting a direct interaction or mCRP expression by the tumor itself.

Conclusion: Our data show that the pro-inflammatory mCRP isoform is expressed in the TME of CC, primarily in patients with high systemic pCRP values. This strengthens the hypothesis that CRP might not only be an inflammatory marker but also an active mediator within tumors.

Keywords: C-reactive protein (CRP); CRP isoforms; biomarkers; colon cancer; immunohistochemistry (IHC); monomeric CRP; systemic inflammation; tumor microenvironment.

Figure 1

Automated Image Analysis Workflow. Left: Whole slide image with annotated tumor regions. Tumor in red and adjacent normal colon mucosa in green. Right: An AI-based algorithm was developed for analyzing the pattern of mCRP distribution and accurately segment tumor epithelium (red) and tumor stroma (blue).

Figure 2

mCRP expression in systemically inflamed and non-inflamed colon cancer patients and adjacent normal colon mucosa. Representative images from patients with (A) normal and (B) elevated circulating CRP. (C) Normal colon mucosa adjacent to the tumor with no mCRP expression. (D) Quantified mCRP (proportion of area with positive mCRP staining) assessed within the tumor and adjacent normal colon mucosa (control) in CRP-high and CRP-low patients.

Figure 3

Prognostic value of tumor mCRP expression and serum CRP in colon cancer patients. (A) A receiver operating characteristics (ROC) curve was calculated to determine the optimal tumor mCRP cutoff value (marked by a bullet) defined as the point on the curve with sensitivity and specificity closest to 100%, corresponding graphically to the point with the minimum distance to the upper left corner (B) Risk of colon cancer death or recurrence above and below the optimal tumor mCRP cutoff value identified from the ROC curve. (C) Risk of colon cancer death or recurrence in CRP-high (serum CRP >30 mg/L) and CRP-low (serum CRP 0-1 mg/L) patients. The optimal mCRP cutoff value was defined.

Figure 4

Correlating mCRP and selected adaptive and innate immune markers in colon cancer patients. Heatmap and corresponding table of Spearman correlations between mCRP and individual immune markers. Red color indicates positive correlation, blue indicates negative correlation, white indicates no correlation.

Figure 5

Colocalization of mCRP with various components of the TME. Representative images from CC patients with elevated serum CRP and pronounced mCRP tumor expression. (A) Highly neutrophil infiltrated tumor area with strong mCRP expression. (B) Necrotic area within a tumor with high mCRP expression. (C) Colocalization of mCRP and macrophages. (D) Colocalization of mCRP and endothelial cells lining intratumoral vessels as well as some mCRP within the vessel lumen. (E) mCRP scattered diffusely as small granules within the connective tissue of the tumor stroma. (F) Tumor cell nuclei surrounded by mCRP (marked by arrows).

Figure 6

Double immunofluorescence labeling of mCRP and tumor cells in colon cancer tissue. Left and middle panels: Unmixed images showing individual stains of mCRP (yellow) to the left and pan-CK positive tumor cells (teal) in the middle. Right panel: Composite image showing double positive mCRP+/pan-CK+ tumor cells (marked by arrows). DAPI (blue) was used for visualization of nuclei. Pan-CK, Pan-cytokeratin.