Purpose:The purpose is to study the prognostic significance of tissue expression of trypsinogen-1, trypsinogen-2, and tumor-associated trypsin inhibitor (TATI) and serum concentration of trypsinogen-2, trypsin-2-API (complex of trypsin-2 with ␣-1-proteinase inhibitor), and TATI in epithelial ovarian cancer.Experimental Design: Expression of trypsinogen-1, trypsinogen-2, and TATI was determined by immunohistochemistry with monoclonal antibodies in tissue sections of tumors from 119 patients with untreated primary epithelial ovarian cancer. Preoperative serum concentrations of trypsinogen-2, trypsin-2-API and TATI were analyzed using specific immunoassays.Results: Fifty-four percent of the tumors expressed trypsinogen-1, 45% expressed trypsinogen-2, and 30% expressed TATI. In patients with stage III and IV disease, TATI tissue expression (P ‫؍‬ 0.002) and elevated TATI concentration in serum (P ‫؍‬ 0.048) were associated with adverse cancer-specific and progression-free survival in univariate analysis. In multivariate analysis, TATI tissue expression (P ‫؍‬ 0.005), tumor grade (P ‫؍‬ 0.0001), histological type (P ‫؍‬ 0.02), and stage (P ‫؍‬ 0.0005) were independent prognostic factors for adverse cancer-specific survival and TATI tissue expression (P ‫؍‬ 0.006) and grade (P ‫؍‬ 0.0003) for progression-free survival. In multivariate analysis of all patients and those with advanced disease, serum trypsin-2-API concentration was an adverse prognostic factor for cancer-specific and progression-free survival, and it was independent of stage and histological type of the tumor (P < 0.01).Conclusions: Tissue expression of TATI and an elevated preoperative serum concentration of trypsin-2-API are strong independent prognostic factors in advanced epithelial ovarian cancer. These results suggest that trypsin expression plays a role in the progression of ovarian cancer. TATI and trypsin-2-API are of potential use as an aid for stratification of randomized studies and for selecting treatment strategies.

Objective To evaluate the use of the pre-operative tumour-associated trypsin inhibitor (TATI) level and residual tumour size at primary surgery as a prognostic indicators for patients with Stage 111 epithelial ovarian cancer.Design Retrospective cohort study. Main outcome measures Cumulative five-year survival, pre-operative serum TATI level and residual tumour size.Results Surgery was optimal (residual tumour size I 2 cm) in 55 patients and suboptimal (residual tumour size > 2 cm) in 43. Pre-operative TATI level I 22 pg/L predicted better prognosis both in patients with optimal and suboptimal surgery compared with patients with pre-operative TATI level > 22 p a . Patients with optimal surgery and a pre-operative TATI > 22 pg/L had a twofold relative risk of death compared with those with a pre-operative TATI I 22 pg/L. The cumulative survival was less than three years for patients with suboptimal surgery and pre-operative TATI > 22 pg/L.Conclusions Pre-operative serum TATI in combination with residual tumour size may be useful in stratifying patients with Stage I11 ovarian cancer into different categories in randomised treatment trials.

In spite of a gradual improvement, survival in epithelial ovarian cancer is disappointingly low. New therapeutic regimens are emerging, and it would be important to be able to predict the prognosis and to stratify patients for clinical trials before therapy. We have evaluated the prognostic value of the pretreatment serum concentrations of 3 tumor markers. The free beta subunit of human chorionic gonadotropin (hCGbeta), CA125 and tumor-associated trypsin inhibitor (TATI) were measured in pretreatment serum samples from 146 patients treated for ovarian cancer between 1990-1995. The patients were followed up until 1998. Elevated concentrations of hCGbeta, CA125 and TATI were observed in 29%, 79% and 33%, respectively. When tested as single variables in Cox's proportional hazards model, stage, grade, size of residual tumor and hCGbeta (all p < 0.001) and CA125 (p = 0.004) correlated with prognosis. However, when fitted as multiple variables together with stage, grade and age in the same model, hCGbeta (RR = 3.42) stage (RR = 2.77) and grade (RR = 3.80) were the only significant variables. When serum hCGbeta was normal, 5-year survival was 80%, but it was only 22% when hCGbeta was elevated. In patients with stage III or IV and minimal residual disease, 5-year survival was 75% if hCGbeta was normal compared with 0% if hCGbeta was elevated. hCGbeta in serum is a strong independent prognostic factor in epithelial ovarian cancer, and its prognostic value is similar to that of grade and stage. The availability of this marker before surgery could facilitate selection of treatment modalities.

No abstract

Serous ovarian carcinoma comprises a clinically heterogenous group of tumors, and molecular markers stratifying patients into clinically meaningful subgroups are needed. Numerous markers have been evaluated, but none of them has yet been routinely incorporated into clinical practice. Previously we have found that elevated serum levels of the free beta subunit of human chorionic gonadotropin (hCGb) and aberrant p53 expression confer poor prognosis in ovarian carcinoma. The aim of our study was to evaluate their combined effect in predicting the outcome of patients with serous ovarian carcinoma. The study material consisted of 173 consecutive patients treated for primary serous ovarian carcinoma in 1 institution between 1990 and 2000. The preoperative serum level of hCGb was analyzed by a ultrasensitive and specific immunofluorometric assay, and p53 tumor tissue expression by immunohistochemistry using a novel classification. Elevated serum hCGb (≥2.0 pmol/L) was detected in 57 (33%) of 173 patients, and aberrant p53 expression in 103 (62%) of 167 interpretable cancers. Elevated hCGb and aberrant p53 expression were strongly associated with poor prognosis (p < 0.0001 for both). Their additive prognostic effect was marked. Five-year survival was 14% (0-29%) when both markers were aberrant, 44% (29-60%) when either one was aberrant and 82% (70-94%) when both were normal. Preoperative serum hCGb and tumor tissue p53 expression are feasible markers that divide serous ovarian carcinomas into clinically relevant subgroups. ' 2007 Wiley-Liss, Inc.