Endometriosis-associated ovarian cancer comprises a special group of ovarian cancers that most probably originate from endometriosis foci. Several in vitro studies have shown that microRNA (miRNA) plays an important role in this carcinogenesis. Our goal was to establish if a distinct miRNA profile can be associated with endometriosis and endometriosis-associated ovarian cancer with their potential causal relationship, and whether such a profile could be used clinically to prognose carcinogenesis in endometriosis foci. We conducted a systematic search according to PRISMA guidelines, registered at PROSPERO (number CRD42021245606). The search encompassed whole Pubmed, Cochrane and Medline databases to 1 May 2025 and the search strategy included the following MeSH terms: ‘miRNAs’ or ‘microRNAs’ or ‘miR’ and ‘ovarian cancer’ and ‘endometriosis’. Our ultimate inclusion criterion was that studies must simultaneously evaluate miRNA expression in endometriosis, regardless of its form and stage, and in endometriosis-associated ovarian cancer (EAOC), as only data generated under identical experimental conditions and using the same controls are truly comparable. The quality of the data was assessed using The Newcastle-Ottawa scale (NOS) and ROBINS-I tool. Our final analysis included 13 studies, comprising 608 patients and over 1000 miRNA molecules. Among those only five manuscripts presented raw data for each miRNA studied. Although several authors declared high sensitivity and specificity for one or more miRNA in distinguishing between endometriosis and endometriosis-associated ovarian cancer, a meta-analysis could not be performed due to the high heterogeneity of the studied samples. We concluded that there is not enough publicly available raw data to establish a set of miRNAs capable of differentiating between the two diseases and of prognosing carcinogenesis. The greatest limitation lies in the use of various standardized reference gene sets, which makes it impossible to compare relative miRNA expression across studies. New data from the next generation sequencing (NGS) experiments would overcome issues related to reference and control genes.
Szubert et al. (Fri,) studied this question.