Endometriosis is an estrogen dependent multi-factorial disease which affects around 10% of women of reproductive age. It is defined as the presence of endometrium-like tissue in sites outside the uterine cavity. Notwithstanding, the etiology and pathogenesis of endometriosis remain uncertain. Combinations of genetic, hormonal, environmental and immunological factors play a role in the pathogenesis of this disorder. Until today, no semi- or non-invasive test has been developed for the diagnosis of endometriosis. The gold standard for the diagnosis of pelvic disease is surgical assessment by laparoscopy. The most important goal of a non-invasive diagnostic test is to identify women with endometriosis who might benefit from surgical treatment for endometriosis-associated pain or subfertility.The overall aim of this thesis was firstly to investigate the pathogenesis of endometriosis using macroscopically normal peritoneum and eutopic endometrium from women with and without endometriosis and secondly, to discover new biomarkers in order to develop a semi- or non-invasive diagnostic test for endometriosis, using endometrium and plasma samples from women with and without endometriosis.We found increased IL-6 mRNA and reduced IL-12 mRNA expression in macroscopically normal peritoneum. This altered gene expression may concurrently contribute to the pathogenesis of endometriosis via enhanced inflammation and via a reduction of natural killer (NK) cytotoxicity. The reduced ferritin mRNA expression in macroscopically normal peritoneum from women with endometriosis suggests that iron overload may be limited to endometriosis lesions and not extend to normal peritoneum. This study indicates that the immunobiology of macroscopically normal peritoneum is relevant to understand the pathogenesis of endometriosis.Proteomic Surface Enhanced Laser Desorption/Ionisation (SELDI-TOF) mass spectrometry (MS) analysis of plasma samples allowed the diagnosis of endometriosis using 5 protein or peptide peaks with high sensitivity (minimal-mild=75%, moderate-severe=98%) and high specificity (minimal-mild=86%, moderate-severe=81%), based on the analysis of a training and test set using a randomization approach. The peak with the highest intensity (2.189Da) was decreased in women with moderate-severe endometriosis when compared to controls and was identified as fibrinogen beta chain peptide. In our combined endometrium microarray and SELDI-TOF MS analysis we showed a clear difference in gene expression in menstrual phase compared to early luteal phase in patients with and without endometriosis. In our endometrium proteomics part of the study we were able to classify minimal-mild versus control using 5 protein or peptide peaks with a sensitivity of 94% and specificity of 100% and when combining minimal-severe with sensitivity of 91% and specificity of 80%. In this dissertation we propose a semi- and non-invasive way to diagnose endometriosis with a high sensitivity and high specificity.