A critical appraisal of the circulating levels of differentially expressed microRNA in endometriosis ^†

Abstract

Endometriosis is a common gynecological condition characterized by estrogen dependence, chronic pelvic pain, infertility, and diagnostic delay of between 5.4 and 12 years. Despite extensive study, no biomarker, either alone or in combination with other markers, has proven superior to laparoscopy for the diagnosis of endometriosis. Recent studies report that circulating levels of differentially expressed microRNA (miRNA) in women with endometriosis compared with controls are potential diagnostic tools. However, the lack of replication and absence of validated differential expression in novel study populations have led some to question the diagnostic value of miRNA. To elucidate potential reasons for the lack of replication of study results and explore future directions to enhance replicability of circulating miRNA results, we carried out an electronic search of the miRNA literature published between 2000 and 2020. Eighteen studies were identified in which 63 different miRNAs were differentially expressed in the circulation of women with endometriosis compared with controls. However, the differential expressions of only 14 miRNAs were duplicated in one or more studies. While individual miRNAs lacked diagnostic value, miRNA panels yielded sensitivity and specificity equal to or better than laparoscopy in five studies. Important differences in study design, sample processing, and analytical methods were identified rendering direct comparisons across studies problematic and could account for the lack of reproducibility of study results. We conclude that while the results of miRNA studies to date are encouraging, refinements to study design and analytical methods should enhance the reliability of circulating miRNA for the diagnosis of endometriosis.

Abstract

miRNA panels in five separate studies were comparable to or better than laparoscopy for the diagnosis of endometriosis; however, important study design and methodological issues account for the lack of replication of results across studies.