Serum Circ‐FAF1/Circ‐ELP3: A novel potential biomarker for breast cancer diagnosis

Abstract Background Recently, measurement of serum circular RNAs (circRNAs) as a non‐invasive tumor marker has been considered more. We designed the present study to investigate the diagnostic efficiency of serum Circ‐ELP3 and Circ‐FAF1, separately and simultaneously, for diagnosis of patients with breast cancer. Methods Seventy‐eight female patients diagnosed as primary breast cancer participated in this study. We measured the level of circRNAs in serum specimens of the studied subjects. A receiver operating characteristic (ROC) curve was plotted and the diagnostic efficiency for both circRNAs was determined. Results Compared to non‐cancerous controls, Circ‐ELP3 was upregulated in breast cancer patients (p‐value = 0.004). On the other hand, serum Circ‐FAF1 was seen to be decreased in breast cancer patients than controls (p‐value = 0.001). According to ROC curve results, the area under the curve (AUC) for Circ‐ELP3 and Circ‐FAF1 was 0.733 and 0.787, respectively. Furthermore, the calculated sensitivity and specificity for Circ‐ELP3 and Circ‐FAF1 were 65, 64% and 77, 74%, respectively. Merging both circRNAs increased the diagnostic efficiency, with a better AUC, sensitivity and specificity values of 0.891, 96 and 62%, respectively. Conclusion Briefly, our results revealed the high diagnostic value for combined circRNAs panel, including Circ‐ELP3 and Circ‐FAF1 as a non‐invasive marker, in detection of breast carcinomas.


| INTRODUC TI ON
Breast cancer is the most frequent cancer diagnosed worldwide according to statistics released by the International Agency for Research on Cancer (IARC) in December 2020. 1 It is expected that there will be 281,550 new cases and 43,600 deaths due to breast cancer within the United States in 2021. 2 An early diagnosis of breast cancer leads to a successful treatment and, therefore, a great chance of survival. Currently, imaging techniques and measurement of serum tumor markers are utilized for screening of breast cancer patients, although both of them have several limitation including high cost and low diagnostic value. 3 Although the sensitivity reported for mammography is between 54% and 77%, this technique is the main tool for breast cancer screening. 3 Histopathological assessment of breast tissue is the gold standard method to confirm the existence of cancer. 4 Besides, several laboratory markers have been approved to be used in breast cancer screening and monitoring, including cancer antigen 15-3 (CA 15-3), carcinoembryonic antigen (CEA) and tissue polypeptide-specific antigen (TPS). However, they are not reliable enough for breast cancer diagnosis, and recent studies showed contradictory results for utilizing these tumor markers. 5 Therefore, there is more attention about finding circulating biomarkers as a reliable tool for clinical management of breast cancer. 6 Among them, recent studies have focused more on the application of molecular biomarkers such as non-coding RNAs (ncRNAs) because of their high specificity and sensitivity. A new subtype of RNAs is circular RNAs, single-strand RNA molecules with less than 100 to more than 4,000 nucleotides 7 and a covalently closed-loop structure. These molecules are produced through a backsplicing mechanism, in which the downstream 5′-end of the splice donor joins the upstream 3′-end splice acceptor and forms a product with a circular structure. 8,9 In a new classification approach, circRNAs can be divided into two categories including coding and non-coding circRNAs. Accordingly, coding circRNAs have several elements such as internal ribosome entry site (IRES), an open reading frame (ORF) and specific m6A site which let them to be translated to mRNAs. 10 Previous studies showed that CircRNAs could involve in various aspects of tumorigenesis like metastasis, invasion, and tumor growth [11][12][13] and, thus, may be considered as a reliable prognostic and diagnostic marker.
It has been shown that circRNAs are more stable to RNase activity compared to linear form. 14,15 Besides, due to their long-time durability in serum, high expression and specificity, circRNAs are considered as favorable biomarkers for diagnosis of various diseases. 16 Previous studies clearly showed that several circRNAs might act as oncogenes in cancer development such as hsa_circ_0001982 17 or circGFRA1. 18 More interestingly, there are some studies that considered circRNAs as a biomarker for cancer management including hsa_circ_0001785 19 and hsa_circ_100219. 20 Experimental analysis on hsa_circ_0001785 and hsa_circ_100219 showed a significant alteration in breast tumors for these molecules and, therefore, introduced them as a possible target for treatment or diagnosis of breast cancer. 19 The gene that encodes hsa_circ_0001785 is elongator complex protein 3 or ELP3, a subunit of the acetyltransferase elongator enzyme complex, which is an associated factor with the RNA polymerase II. 21 Previous studies indicated a significant elevation in ELP3 expression in breast tumors. It has been suggested that ELP3 could enhance breast cancer metastasis via its role on the wobble uridine (U34) of tRNA modification. 22 FAF1 protein is a potent inhibitor of the TGFβ signaling pathway. FAF1 overexpression can reduce the metastasis and invasion of breast tumors; thereby, downregulation of FAF1 has a close correlation with increased metastasis in breast cancer. 23 It was found that hsa_circ_100219 produces from FAF1 and high level of this circRNA in breast cancer patients can remarkably suppress the proliferation, cell migration, and invasion of cancer cells. Also, hsa_circ_100219, through acting as a miR-942 sponge, can upregulate the expression of suppressor of cytokine signaling 3 (SOCS3). 24 Despite the established role of hsa_circ_100219 and hsa_ circ_0001785 in breast cancer development, the possible use of these two circRNAs in the clinic is still unclear. Therefore, the present study was designed to investigate the diagnostic value of hsa_ circ_0001785 (Circ-ELP3) and hsa_circ_100219 (Circ-FAF1) in serum samples of breast cancer patients before and after an intervention to find out whether these circRNAs can utilize as a diagnostic and prognostic biomarker for human breast cancer assessment. by imaging approaches. This study was approved by the Regional

Committee of Ethics of the Kurdistan University of Medical
Sciences. For staging and grading the patients, Scarf-Bloom-Richardson criteria and TNM staging system were applied. 4,25 All clinical, laboratory and pathological details were obtained from patients' medical records.

| Sample collection
For specimen collection, 5 ml whole blood was collected from patients before any therapeutic intervention. Six months after beginning treatment (mastectomy, chemotherapy, radiotherapy, or a combination of all), another whole blood sample was obtained from patients. At the same time, a single blood sample was obtained from non-cancerous subjects. For serum separation, centrifugation was performed at 3500 rpm for 5 min. Subsequently, each separated serum was aliquoted in two vials and was stored at −80°C upon the analysis. 26-28

| Quantitative real-time PCR analysis
According to the manufacturer's protocol, total RNA isolation from serum samples was performed using a Sansure Mag kit (Sansure Biotech, China). Eventually, the quality and quantity of isolated

| Statistical analysis
Data analyzing performed by SPSS 16 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 8.2.1 (GraphPad Prism Inc., San Diego, CA, USA). We used mean ± standard deviation (SD) for representing the results. Then, for data comparison between the mean of the studied subjects, Mann-Whitney test and one-way ANOVA analysis were performed, and p-values <0.05 were considered as statistically significant values. Using a receiver operating characteristic (ROC) curve, cutoff values were determined and then the sensitivity and specificity for each circRNAs were calculated.

| Clinical characteristics of study subjects
The mean ages for patients and control group were 46.42 ± 10.94 and 43.21 ± 6.71, respectively (p-value = 0.28). Among cases, 63.6% were positive for human epidermal growth factor receptor-2 (HER2), and 36.4% were negative. Our results showed that among studied subjects, 71.9% were positive for estrogen receptor (ER) while 28.1% were ER negative. Also, we found that 32.8% of patients were negative for progesterone receptor (PR), and 67.2% were PR posi-

| Expression levels of the studied circRNAs
The expression level of two studied circRNAs hsa_circ_0001785 (Circ-ELP3) and hsa_circ_100219 (Circ-FAF1) is illustrated in Figure 1. Our results showed that the circulating level of hsa_ circ_0001785 (Circ-ELP3) in breast cancer patients before treatment was upregulated compared with controls (p-value = 0.0106), while after treatment, the level of this circRNA was significantly decreased compared to pre-treatment status (p-value = 0.01) and, moreover, this value had no statistically significant difference with control group (p-value = 0.9451) ( Figure 1A). As illustrated in Figure 1B, the circulating level of hsa_circ_100219 (Circ-FAF1) in serum specimen of patients before treatment was significantly lower than controls (p < 0.0001), while the expression level showed a statistically significant overexpression after treatment (p-value = 0.0069). Additionally, we evaluated the re-

| Diagnostic value of studied circRNAs
By drawing a ROC curve, we determined the diagnostic values of hsa_circ_0001785 (Circ-ELP3) and hsa_circ_100219 (Circ-FAF1) for diagnosis of breast cancer (Figure 2A Table 4; as mentioned in this table, Circ-FAF1 has higher diagnostic efficiency for breast cancer detection according to the AUC value.
Finally, we combined the two circRNAs to see whether it produce a better diagnostic value for detection of breast cancer. As it is shown in Figure 2C, the combination of them showed higher AUC.
Furthermore, using a combined panel, the sensitivity of the test was highly increased and showed higher diagnostic efficiency for breast cancer patients compared to a single test panel (Table 4). The diagnostic value for CEA and CA15-3 has been studied well.

| DISCUSS ION
In a recent study by Uygur et al., 30 measuring serum CEA and CA 15-3 had shown highest sensitivity for hormone receptor and highest specificity for HER2 status (88.17% and 60%, respectively). In another study by Wand et al., 31 CEA and CA15-3 showed low sensitivity (56.7% and 44.5%, respectively) and high specificity (92% and 84.5%, respectively) for diagnosis of metastatic breast cancers.
A recent meta-analysis about the diagnostic efficacy of CEA and CA15-3 in patients with breast cancer revealed that higher plasma CEA and CA15-3 are correlated with poor disease-free survival and overall survival, and therefore, suggested that they might be evaluated anytime if possible. 32 Collectively, utilizing these tumor markers in breast cancer diagnosis is still controversial. Besides, various imaging methods have several limitations such as high cost and low sensitivity or specificity. 29 Therefore, more studies have focused on introducing novel circulating tumor markers for breast cancer diagnosis and monitoring of treatment.

ACK N OWLED G M ENTS
The author wish to thank all patients and health stuffs who participated in this study. Financial support from Kurdistan University of medical sciences is highly appreciated.

CO N FLI C T O F I NTE R E S T
Mr. R Omid-Shafa'at declares no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

I N FO R M E D CO N S E NT
Informed consent was obtained from all individual participants included in the study.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.