Whole blood-derived miRNA profiles as potential new tools for ovarian cancer screening

Patients suffering from ovarian cancer (OvCA) are still burdened by the most unfavourable prognosis of all gynaecological cancers (Pectasides and Pectasides, 2006). This is largely due to the generally late detection of the disease: while 5-year survival is 90% in those 25% of cases in which diagnosis is achieved at FIGO I stage (Duffy et al, 2005; Badgwell and Bast, 2007), long-term survival becomes very limited at advanced stages FIGO III and IV (combined 5-year survival rate ∼10%) (Duffy et al, 2005). Accordingly, there is a major interest in the discovery of biomarkers for the early detection of OvCA (Clarke-Pearson, 2009). However, even CA125 that was the most promising single marker found in serum is neither sensitive nor specific enough (Meany et al, 2009) and therefore not recommended for screening of asymptomatic women (Duffy et al, 2005). Sensitivity is biologically limited by the lack of CA125 (over)expression in approximately 50% of OvCAs at FIGO stage I (Jacobs and Bast, 1989). Specificity is also a problem because approximately 1% of all healthy women seem to have elevated levels of this marker (Bast et al, 1983). In addition, several benign conditions such as endometriosis, pelvic inflammations, ovarian cysts or even pregnancy (Duffy et al, 2005) also result in increased CA125 levels. As a tool for the monitoring of OvCA recurrence, CA125 is also of very limited use. Until now there is no evidence that an earlier initiation of suitable therapies on increases in CA125 levels translates into a prolonged survival (Eisenhauer et al, 1997). Thus, surveillance of OvCA patients with CA125 is not recommended at the moment (Duffy et al, 2005). Considering that all efforts to identify suitable protein biomarkers were largely futile, we turned our attention to microRNAs (miRNAs). These small (17–24 nucleotides) non-coding RNAs (Lee et al, 1993) regulate many physio- and pathological processes through control of gene expression (Calin and Croce, 2006; Zhang et al, 2007). As opposed to mRNAs, miRNAs are active moieties by themselves and should thus reflect physiological alterations more directly (Gilad et al, 2008). A de-regulation of miRNA expression has already been described in numerous malignancies including OvCA in which it was functionally connected to the inhibition of apoptosis (Yang et al, 2008; Zhang et al, 2008). As tumour-associated miRNA patterns are highly tissue-specific, they can allow an identification of the origin of tumour metastases (Rosenfeld et al, 2008). Moreover, miRNAs are also remarkably stable which allows their easy isolation and analysis from tissues and from blood in which they can be found both as free circulating nucleic acids and in mononuclear cells (Chen et al, 2008). The possibility to analyse multiple miRNAs in parallel through nucleotide arrays further offers the possibility to increase sensitivity and specificity by using complex miRNA expression patterns as opposed to single biomarkers. Thus, miRNAs might constitute very useful and accessible diagnostic tools (Chen et al, 2008; Gilad et al, 2008). Accordingly, we used the latest and most complete collection of miRNA sequences analysed to date to identify potential differences between the blood-derived miRNA profiles of OvCA patients and healthy volunteers. On the basis of the findings of our proof-of-principle study, we suggest that this new approach holds considerable promise for the development of improved screening and surveillance strategies for OvCA.