miRNAs in Sera of Tunisian patients discriminate between inflammatory breast cancer and non-inflammatory breast cancer

Breast cancer (BC) is the most commonly diagnosed cancer in women worldwide. In Tunisia, this tumor represents the leading cause of cancer death among women, accounting for 30% of all malignancies in women and with a particularly high proportion of inflammatory types of breast cancer (Belkacémi et al. 2010).

Inflammatory breast cancer (IBC) is rare but the most lethal form of locally advanced breast carcinoma, occurring in 1% to 6% of all invasive breast tumors. (Lehman et al. 2012); affected patients have a poor prognosis due to infiltration of tumor emboli that metastasize within the dermal lymphatic vessels of the skin overlying the breast (Boussen et al. 2010). According to the American Joint Committee on Cancer (AJCC), IBC is “a clinico-pathologic entity characterized by diffuse erythema and edema (i.e., peau d’orange) of the breast, often without an underlying palpable mass” (Lehman et al. 2012). Furthermore, the AJCC’s staging scheme defines IBC as T4d tumor at diagnosis, and characterizes it as typically affecting young women, often during their childbearing years (Laere et al. 2005). At diagnosis, a majority of patients have axillary lymph nodes involved, and one third present with distant metastases (Bertucci et al. 2005).

Gene expression profiles provide specific molecular signatures containing information which may enable explanation of the mechanisms of tumor development and progression and the poor prognosis observed in IBC. Indeed, several studies comparing IBC to non-IBC have demonstrated a specific mRNA expression signature in IBC that is significantly enriched for genes involved in cell motility, inflammation, immune response and stem cell biology (Bertucci et al. 2004; Ben Hamida et al. 2008). Furthermore, expression profiling performed by Marrahchi et al. have further strengthened the evidence of the role of WISP3 as a tumor suppressor gene and RhoC as an oncogene which act in concert to give rise to the highly aggressive IBC phenotype in the Tunisian population. (Marrakchi et al. 2009).

Some studies have noted the dysregulation of several miRNA in IBC tumors, but (Iorio et al. 2005; Blenkiron et al. 2007; Auwera et al. 2010), no study has focused on miRNA levels in sera of patients with IBC. The aim of this study was to identify a molecular signature of IBC based on miRNA profile in serum and to use this profile as an early diagnosis criterion. To this end, we present the first investigation of a panel of miRNA in the serum of inflammatory breast cancer patients and compared this profile to non-inflammatory breast cancer cases and non-cancer controls.

Tumor tissue miRNAs have been demonstrated to be associated with the pathogenesis of breast cancer and proposed to be used as biomarkers for diagnosis and prognosis (Lowery et al. 2008); (Iorio et al. 2005); (Blenkiron et al. 2007). (Giglio et al. 2013). At least 24 candidate oncogenes with gains at the DNA level, discovered with comparative genomic hybridization (CGH) studies, have been reported for IBC in contrast to non-IBC breast tumors. Gene expression studies also support genetic heterogeneity for IBC (Bekhouche et al. 2011).

Although notable progress has been made in the last decade in the study of IBC, early detection is one of the major challenges (Lehman et al. 2012). Within this context, microRNAs (miRNAs), which are non-coding, single-stranded small RNAs, offer a promising new avenue of research since they are responsible for regulating gene expression on the post-transcriptional level. They accomplish this regulation through binding, due to partial or total sequence homology to target mRNAs, thus inhibiting the translation of protein from these mRNAs or by promoting their degradation (Wu et al. 2009). Through this process, miRNAs are involved in the regulation differentiation, metabolism, proliferation and apoptosis (Chen et al. 2012). It has been observed that dysregulated miRNA expression resulted in a wide array of pathological events, including cancer, indicating that miRNAs may become valuable molecular markers for detection and prognosis of tumors. Compared to protein-based biomarkers, miRNA offers several advantages: the complexity of miRNA is much lower than that of proteins; the sequences of many miRNAs are conserved across species; the expression of some miRNAs are restricted to specific tissues or biological stages ; and the levels of miRNAs can also be easily measured by various commonly used laboratory methods (Friedman et al. 2012).

More recently, Mitchell et al., reported that miRNAs existed in serum/plasma, and were originated from cancer tissue (Mitchell et al. 2008). The extreme stability of circulating miRNAs appeared due to their resistance to RNase digestion and to their protection from degradation through inclusion in various protein complexes or membranous particles such as exosomes or microvesicles. The miRNAs are particularly resistant to harsh conditions such as extended storage and multiple freeze-thaw cycles which holds great promise for the use of miRNAs as distinctive, non-invasive cancer biomarkers (Ng et al. 2013);(Blondal et al. 2012).

Since the discovery of circulating miRNAs, numerous studies on miRNA and cancer have been published (Volinia et al. 2006). Focusing on breast cancer, a pilot study performed by Roth C et al. provided the first evidence that tumor-associated circulating microRNAs are elevated in the blood of breast cancer patients and associated with tumor progression (Roth et al. 2010).

Moreover, Heneghan et al. found that expression of miR-195 was significantly elevated in breast cancer patients compared to healthy controls (Heneghan et al. 2010).

The role of serum miRNA in breast cancer has not been thoroughly studied. Our work is based on the hypothesis that circulating miRNAs in serum can potentially serve as novel, minimally invasive biomarkers for early detection of inflammatory and non-inflammatory breast cancer. In the present study, we investigated the potential of 12 circulating miRNAs, selected from previous studies, by analyzing serum profiles of IBC, non-IBC and matched healthy controls. The clustering of miRNA expression profiles that we observed identified four groups of miRNAs that sufficiently discriminated the three groups of patient sera in a blinded fashion.

Moreover, normal sera profiles are rather related to low levels of miRNAs which is in agreement with the fact that most of the miRNAs studied could be considered oncogenes (Du et al. 2013) (Li et al. 2010). However two profiles with high miRNA expression and close to some non-IBC patient sera, are observed in controls leading to a doubt on their normality. In other words such profiles could be associated to the first steps of the disease. Since no clear clusters corresponding to the three serum groups have been carried out, this hypothesis needs more investigation to be confirmed.

Moreover, our results have shown that taken individually some serological mRNA could be pertinent markers of breast cancer. Indeed, we observed that, from the 12 miRNAs studied, five miRNAs (mir-24, mir-342-3p, mir-30b, mir-337-5p and mir-451) were significantly dysregulated in the serum of the 3 groups studied, among which four discriminated between IBC and non-IBC patient sera.

Although serum samples are easily acquired in a relatively noninvasive manner, and isolated miRNAs are readily detected by qRT-PCR, a widely used clinical and laboratory technique, it should be mentioned that miRNA profiles in serum are not necessarily representative of their expression in the tumor. Hence, our results revealing only four discriminative miRNAs between IBC and non-IBC sera are discordant with a previous study on breast tumors that observed 3 times as many discriminative miRNAs. Indeed, since miRNA presence is often observed in healthy subject sera, it is obvious that it may be a consequence of normal physiology. In breast cancer patient serum, this serological expression may be viewed as a cause or a consequence of the oncogenic process. Therefore, if it can be shown prospectively that the levels are over- or under-expressed prior to cancer diagnosis, the determination of the serum miRNA profiles in breast cancer patients may be quite meaningful. In our study, mir-342-3p and mir-24were significantly up-regulated in non-IBC as compared with IBC. This result is in agreement with a previous report on IBC and non-IBC miRNAs expression in tumors, while our results differed in showing a significant up-regulation of miR-337-5p in IBC (Iorio et al. 2005). This up-regulation phenomenon may also be the case for miR-30; moreover the miR-30 family is able to suppress apoptosis (Li et al. 2011) and is up-regulated in non-IBC tumors in comparison with IBC (Iorio et al. 2005)/(Blenkiron et al. 2007). Finally, we also observed that mir-451 was significantly down-regulated in IBC serum as compared with non-IBC and with healthy controls, with adequate specificity and sensibility to serve as a marker for IBC. The lower expression level of mir-451 observed in serum of IBC comparatively to normal sera could be either the consequence of the tumor development or the cause of the disease through more general mechanisms leading to the oncogenic process in IBC. In the last case, mir-451 low expression could be used as an early diagnosis marker of IBC. Our results are in agreement with previous studies showing mir-451 as a tumor suppressor (Li et al. 2013).

In conclusion, if validated in future, larger studies, mir-451 could be considered as a serological marker for IBC, and mir-337-5p and mir-30 for non-IBC, with clear utility in early diagnosis and/or disease prognosis studies. Therefore the continued study of these miRNA dysregulation pathways may lead to breast cancer prevention strategies.

Jan Blancato and R Marrakchi are senior authors.