Ovarian Cancer-Cell Pericellular Hyaluronan Deposition Negatively Impacts Prognosis of Ovarian Cancer Patients

Abstract

Background: Hyaluronan (HA), a component of the extracellular matrix, is frequently increased under pathological conditions including cancer. Not only stroma cells but also cancer cells themselves synthesize HA, and the interaction of HA with its cognate receptors promotes malignant progression and metastasis.

Methods: In the present study, HA deposition in tissue sections was analyzed by hyaluronan-binding protein (HABP) ligand histochemistry in 17 borderline tumors and 102 primary and 20 recurrent ovarian cancer samples. The intensity and, particularly, localization of the HA deposition were recorded: for the localization, the pericellular deposition around the ovarian cancer cells was distinguished from the deposition within the stromal compartment. These histochemical data were correlated with clinical and pathological parameters. Additionally, within a reduced subgroup of ovarian cancer samples (n = 70), the RNA levels of several HA-associated genes were correlated with the HA localization and intensity.

Results: Both stroma-localized and pericellular tumor-cell-associated HA deposition were observed. Cancer-cell pericellular HA deposition, irrespective of its staining intensity, was significantly associated with malignancy, and in the primary ovarian cancer cohort, it represents an independent unfavorable prognostic marker for overall survival. Furthermore, a significant association between high CD44, HAS2 and HAS3 mRNA levels and a cancer-cell pericellular HA-deposition pattern was noted. In contrast, stromal hyaluronan deposition had no impact on ovarian cancer prognosis.

Conclusions: In conclusion, the site of HA deposition is of prognostic value, but the amount deposited is not. The significant association of only peritumoral cancer-cell HA deposition with high CD44 mRNA expression levels suggests a pivotal role of the CD44-HA signaling axis for malignant progression in ovarian cancer.

Keywords: hyaluronan-related enzymes; ovarian cancer; stromal hyaluronan; tumor-associated hyaluronan staining pattern.

Figure 1

Hyaluronan deposition in ovarian tumors and its association with malignancy. Hyaluronan was detected by ligand histochemistry using a biotinylated hyaluronan-binding protein (HABP) in a patient cohort including 15 borderline tumors, and 102 primary and 20 recurrent ovarian cancer samples. Two staining patterns were detected: fine reticular peri-ovarian cancer-cell HA deposition (A,B) and coarse stromal HA deposition (C,D). A significant correlation between fine reticular peri-ovarian cancer-cell HA deposition and malignant potential could be observed (E). The p value after chi-square tests is provided; p < 0.05 was regarded as significant.

Figure 2

Kaplan–Meier analysis of overall survival based on HABP-staining intensity and staining pattern. (A) Kaplan–Meier analysis showed a significant correlation of pericellular ovarian cancer-cell HA deposition with shorter overall survival. (B) No significant association of HA-deposition intensity with overall survival including all primary tumors could be found. p values after log-rank tests are shown; p < 0.05 was regarded as significant.

Figure 3

Correlations between HA-deposition pattern and expression of HA-related genes. Tumors displaying an ovarian cancer-cell pericellular hyaluronan deposition pattern showed higher mRNA levels of (A) CD44, (B) HAS2 and (C) HAS3. p values after chi-square tests are provided; p < 0.05 was regarded as significant.