Cysteine proteinases cathepsins (Cats) B and L and their endogenous inhibitors stefins (Stefs) A and B are implicated in the processes of local and metastatic tumor spread. They were identified as potential prognosticators in various malignant diseases, particularly in breast cancer. The aim of the present study was to determine the concentrations of Cats B and L and Stefs A and B in the tumor and adjacent normal tissue samples collected from 49 patients (the present group) with squamous cell carcinoma of the head and neck (SCCHN), using quantitative immunosorbent assays (ELISA; KRKA d.d., Novo mesto, Slovenia). Their clinical significance was compared with that from a previous study (the reference group, 45 patients; Budihna et al., Biol. Chem. Hoppe-Seyler, 377: 385-390, 1996). The follow-up of patients from the latter report was updated for this purpose. In the present group, significantly higher concentrations of Cat B (P < 0.0001), Cat L (P < 0.0001) and Stef A (P = 0.006) were found in tumors compared with concentrations in their normal tissue counterparts. Cat concentrations in normal laryngeal tissue were significantly/marginally elevated compared with nonlaryngeal tissue (Cat B, P = 0.02; Cat L, P = 0.06). The tumor concentration of Cat L was found to correlate with pT classification (P = 0.005) and tumor-node-metastasis stage (P = 0.05), whereas the concentrations of Stefs A and B correlated with pN classification (P = 0.007 and P = 0.03, respectively) and tumor-node-metastasis stage of the disease (P = 0.02 and P = 0.03, respectively). There was no statistically significant difference between low and high Cat B or Cat L groups, regarding either disease-free survival or disease-specific survival, using a minimum P approach to determine cutoff concentrations. The risk of disease recurrence and SCCHN-related death was significantly higher in patients with low Stef A (P = 0.0006 and P = 0.0005, respectively) and Stef B (P = 0.0009 and P = 0.0007, respectively) tumors, compared with those with high-Stef A and Stef B tumors. These results remained significant even after Ps were adjusted for a possible bias in the estimated effect on survival. The survival analysis in the reference group also confirmed these findings (Stef A: P = 0.0009 and P = 0.002, respectively; Stef B: P = 0.03 and P = 0.009, respectively). To avoid any possible bias arising from the differences between the laboratories that performed the biochemical analysis, the concentrations of both Stefs in the present group and in the reference group were standardized and coupled together to form a uniform group. In univariate survival analysis, standardized values of Stef A and Stef B correlated inversely with the rate of relapse (P = 0.0000) and mortality rate (P = 0.0000). Multivariate regression analysis showed that the standardized value of Stef A is the strongest independent prognostic factor for both disease-free survival and disease-specific survival. These findings show the specific role of Cats B and L and Stefs A and B in the invasive behavior of SCCHN. Furthermore, Stef A proved to be a reliable prognosticator of the risk of relapse and death in patients with this type of cancer.