Pelvic inflammatory disease and epithelial ovarian tumors

Sammanfattning: Background: Epithelial ovarian cancer and borderline ovarian tumors consist of several histotypes in which high-grade serous carcinoma is the most common. The majority of epithelial ovarian tumors are considered to originate in the fimbriated end of the fallopian tubes. What initiates these tumors is far from completely understood. Pelvic inflammatory disease has been proposed as a modifiable risk factor for epithelial ovarian tumors. A major cause of pelvic inflammatory disease is Chlamydia trachomatis which has been shown to have cancer-causing potential. The overall purpose of this thesis was to study associations of pelvic inflammatory disease and C. trachomatis with risk of epithelial ovarian tumors.Methods: In a cross-sectional study (Paper I) we collected ovarian tissue and corresponding blood samples from 69 women undergoing surgery due to suspected ovarian pathology. C. trachomatis specific protein (immunohistochemistry) and C. trachomatis DNA (qPCR) in ovarian tissue were analyzed (Paper I). In a nested case-control study (Paper II) prospective blood samples from 92 women diagnosed with high-grade serous ovarian cancer were matched to four controls each for age and date of plasma sampling. C. trachomatis specific plasma antibodies were analyzed by commercial Enzyme-Linked ImmunoSorbent Assay (ELISA) and Micro-ImmunoFluorescence (MIF) (Paper I and Paper II). We performed a nationwide register-based case-control study where we included 15 072 women diagnosed with epithelial ovarian cancer (Paper III), 4782 women diagnosed with borderline ovarian tumors (Paper IV), and ten controls each matched for age and residential district. Using national Swedish registers, we retrieved data on historyof pelvic inflammatory disease and the potential confounding factors parity, educational level, previous gynecological surgery, and hormonal therapy.Results: We found C. trachomatis DNA in ovarian tissue of eight women with ovarian carcinoma, but not in ovarian tissue from women with borderline ovarian tumors or benign disease (Paper I). The prevalence of the C. trachomatis specific protein did not differ in benign and malignant tissue (Paper I). Prevalence of C. trachomatis specific plasma antibodies was similar in cases and controls at diagnosis (Paper I) and prospectively (Paper II). A history of clinically verified pelvic inflammatory disease was associated with an increased risk of epithelial ovarian cancer overall (Paper III) and borderline ovarian tumors overall (Paper IV). Histotype-specific analyses showed an increased risk of serous carcinoma (Paper III), high-grade serous carcinoma (Paper III), clear cell carcinoma (Paper III), and serous borderline ovarian tumors (Paper IV) but not significantly with other histotypes. A dose-response relationship was seen between an increased number of pelvic inflammatory disease episodes and epithelial ovarian cancer (Paper III), as well as borderline ovarian tumors (Paper IV).Conclusions: This thesis contributes to an improved understanding of the association between pelvic inflammatory disease and epithelial ovarian tumors. The results regarding C. trachomatis are inconclusive and suggests that the association of pelvic inflammatory disease with epithelial ovarian tumors acts through mechanisms other than Chlamydia alone.