Evaluation of the prognostic potential of hyaluronic acid and hyaluronidase (HYAL1) for prostate cancer

Despite the development of nomograms designed to evaluate the prognosis of a patient with prostate cancer (CaP), the information has been limited to prostate-specific antigen (PSA), clinical stage, Gleason score, and tumor volume estimates. To improve our ability to predict prognosis, information regarding the molecular properties of CaP is needed. Hyaluronic acid (HA) is a glycosaminoglycan that promotes tumor metastasis. Hyaluronidase (HAase) is an enzyme that degrades HA into angiogenic fragments. We recently showed that in CaP tissues, whereas HA is localized mostly in the tumor-associated stroma, HYAL1 type HAase is exclusively localized in CaP cells (Lokeshwar et al. J. Biol. Chem., 276: 11922-11932, 2001). We evaluated the prognostic potential of HA and HYAL1 in CaP by immunohistochemistry. Archival CaP specimens were obtained from patients who underwent radical retropubic prostatectomy for clinically localized CaP. Group 1 (n = 25) included patients who showed biochemical recurrence (PSA >0.4 ng/ml; mean recurrence: 21.3 months). Group 2 included patients with no clinical or biochemical recurrence (n = 45; mean follow-up: 80.9 months). For HA and HYAL1 staining, a biotinylated HA-binding protein and an anti-HYAL1 antibody were used. The staining was evaluated on the basis of intensity (0 to 3+) and as dense or sparse (for HA staining only) and then grouped as low grade and high grade. In CaP specimens, HYAL1 was exclusively expressed in tumor cells. Although the stroma was stained positive for HA, 40% of tumor cells also expressed HA. HA, HYAL1, and combined HA-HYAL1 staining predicted progression with 96%, 84%, and 88% sensitivity, 55.5%, 80%, and 84.4% specificity, and 70%, 81.4%, and 85.7% accuracy, respectively. In the univariate analysis, preoperative PSA, Gleason sum, stage, margin, seminal vesicle, extra-prostatic extension (EPE), HA, HYAL1, and HA-HYAL1 were significant in predicting progression (P < 0.05). However, in the multiple logistic regression analysis, only EPE [odds ratio (OR) = 33.483; P = 0.002), HYAL1 (OR = 12.42; P = 0.009), HA-HYAL1 (OR = 18.048; P = 0.0033), and margin (OR = 26.948; P = 0.006)] were significant. Thus, in this 5-year follow-up study, HYAL1, together with EPE and margin, was found to be an independent prognostic indicator.