Highly expressed circ_0000285 from serum and cervical exfoliated cells as a novel biomarker for the diagnosis of early stage-cervical cancer

Abstract Circ_0000285 is reported to play an oncogenic role in the development of cervical cancer (CC). The aim of this research was to investigate the diagnostic power of circ_0000285 in CC. The expression of circ_0000285 in 116 healthy volunteers, 65 early-stage CC (ESCC) patients, and 87 locally advanced CC (LACC) patients was detected by qRT-PCR. The diagnostic values of circ_0000285 for CC and ESCC were evaluated by ROC curves analysis. The circ_0000285 expression was upregulated in serum and cervical exfoliated cells from preoperative CC patients compared to that of healthy volunteers. Increased circ_0000285 expression was found in preoperative LACC patients more than that in ESCC patients. The circ_0000285 expression was downregulated in serum from CC patients after surgery. The postoperative CC patients with high serum circ_0000285 expression was more prone to have a tumour relapse. High circ_0000285 expression was positively correlated with SCC-Ag level and HPV positive rate. The AUC of circ_0000285 for the diagnosis of CC and ESCC were 0.855 and 0.804, better than CA125 and SCC-Ag. When circ_0000285, CA125, SCC-Ag and HPV were combined, the AUC could reach 0.911 and 0.894. In summary, highly expressed circ_0000285 from serum and cervical exfoliated cells might be a promising diagnostic biomarker for ESCC. Impact statement What is already known on this subject? The CA125 and SCC-Ag have limitations in the diagnosis of ESCC. Recently, circRNAs have caused great attention and have been developing rapidly in clinical diagnosis of malignant tumours. What do the results of this study add? Highly expressed circ_0000285 from serum and cervical exfoliated cells might be used as a novel, non-invasive biomarker for the diagnosis of ESCC. What are the implications of these findings for clinical practice and/or further research? Circ_0000285 is superior to CA125 and SCC-Ag for the diagnosis of ESCC in clinical practice. The results help to supplement the shortcomings of traditional tumour markers and improve the diagnosis of ESCC.


Introduction
Cervical cancer (CC) is one of the most common malignant tumours in women worldwide, which is mainly caused by the continuous infection of high-risk human papillomavirus (HPV) (Hu andMa 2018, Siegel et al. 2022).The early-stage CC (ESCC) often has no clinical signs; thus, approximately more than one-third of patients were diagnosed with locally advanced CC (LACC) when symptoms appear (Di Donna et al. 2022).Despite receiving recommended treatments, patients with LACC have a poor prognosis with the 5-year overall survival of 50-70% (Hill 2020).Elevating early diagnosis is of great significance for improving the prognosis of CC patients.Serum non-invasive diagnostic biomarkers including Carbohydrate antigen 125 (CA125) and Squamous cell carcinoma antigen (SCC-Ag) are widely used biomarkers for the detection of CC in routine clinical practice (Dasari et al. 2015).Unfortunately, there are clear limitations to the use of these biomarkers for the diagnosis of ESCC.Therefore, there is an urgent need to identify novel and effective biomarkers to address this challenge.
With the rapid development of RNA deep sequencing technology and bioinformatics in recent years, more and more circular RNAs (circRNAs) have been identified and paid more attention.The circRNAs are a type of non-coding RNAs with a continuous covalent closed-loop structure, which are abundant, conserved, and stable in mammalian cells, and play an essential role in various cellular pathological and physiological processes, including tumorigenesis and metastasis (Kristensen et al. 2019;Patop et al. 2019;Shen et al. 2019).Since circRNAs are not easily degraded by nucleases, making it is used as promising and technically suitable biomarkers for the detection of human diseases (Wang et al. 2021;Zhang et al. 2018).Actually, several circRNAs, such as circ_TOLLIP, circ_0104824, circ_0141633, and circ_0067934, have shown their potential as non-invasive biomarkers for lung cancer (Peng et al. 2021), breast cancer (Li et al. 2020), gastric cancer (Yin et al. 2020), and hepatocellular carcinoma (Zhou et al. 2020), respectively.However, there are seldom circRNAs reported to be a biomarker in the diagnosis of ESCC.
The circRNA_0000285 (circ_0000285) is formed by circularisation of exons 7 and 8 of Homeodomain interacting protein kinase 3 (HIPK3) transcript (NM_005734), with a total of 284 bp.The circ_0000285 is previously identified in CC and exhibited an increased expression in cancer tissues (Chen et al. 2019).Another study provided evidence that circ_ 0000285 plays an oncogenic role in the development of CC by regulating cell proliferation and apoptosis (Zhang et al. 2022).However, it is still unknown whether circ_0000285 could serve as a potential biomarker of CC.In this study, we collected the serum and cervical exfoliated cells from a cohort of healthy volunteers and CC patients and verified the expression levels of circ_0000285.And evaluated its diagnostic power for CC and ESCC.Results showed that highly expressed circ_0000285 could be used as a promising biomarker for the diagnosis of ESCC, better than CA125 and SCC-Ag.The findings of this study will assist the development of screening tests using circ_0000285 from serum and cervical exfoliated cells for the diagnosis of ESCC.

Study subjects
The size of study subjects was calculated using the one-sided 95% formulation of the lower confidence limit according to the previous reference (Flahault et al. 2005).A total of 116 cases of healthy volunteers and 152 cases of CC patients (65 cases of ESCC and 87 cases of LACC) were enrolled at Tianjin Fifth Central Hospital (Tianjin, China) from January 2015 to September 2019.Healthy volunteers were age-matched to the CC patients and free of any known malignancies.All cases of CC were histopathologically confirmed by at least two pathologists.The exclusion criteria for CC were as follows: 1) patients with other types of cancer besides CC; 2) patients with benign cervical lesions, such as severe cervical erosion, cervical tuberculosis, and cervical polyps with microglandular hyperplasia; 3) patients with severe mental disorders; 4) patients who cannot tolerate surgery due to severe liver, kidney, heart and lung insufficiency, coagulation dysfunction, poor control of blood sugar and blood pressure; 5) patients with insufficient liquid-based cytology (LBC) and HPV DNA test.The neoplasm staging was in accordance with the International Federation of Gynaecology and Obstetrics (FIGO) (Saleh et al. 2020), and the ESCC and LACC were defined as stage IA1-IB1 and stage IB2-IVA, respectively.None of the patients had previously undergone chemotherapy, radiation therapy, and immunotherapy.
The detailed clinical characteristics (such as age, body mass index, smoking, drinking, CA125 and SCC-Ag levels, and HPV positive rate) were listed in Table 1.The criteria for smoking: continuous or cumulative smoking for more than 6 months (�10 sticks/d).The criteria for drinking: drinking at least 20 mL/d for more than 1 year.The levels of CA125 and SCC-Ag were analysed through tumour marker detection.All patients were followed up postoperatively for 2 years, the tumour relapse was evaluated based on NCCN clinical practice guidelines in CC (Koh et al. 2019).This study was in accordance with the Enhancing the Quality and Transparency of Health Research (EQUATOR) network guidelines.This research was approved by the ethical review board of Tianjin Fifth Central Hospital (approval no.2018,023) in accordance with the ethical guidelines of the 1975 Declaration of Helsinki.The informed written consent was obtained from all participants.

Samples collection
Fresh peripheral venous blood samples (5 ml each) were collected with ethylenediaminetetraacetic acid-containing tubes (BD Biosciences, Franklin Lakes, NJ, USA) from healthy volunteers and preoperative CC patients.In addition, paired blood samples were obtained from CC patients 90 days after the surgery.The blood samples were centrifuged at 800 � g for 15 min at 4 � C to obtain serum using standard procedures, and the serum was divided into multiple aliquots of 500 mL each in RNase-free tubes (ZYMO Research, Irvine, CA, USA).The cervical exfoliated cells were collected by LBC test after the addition of 500 mL of PBS.All samples were frozen in liquid nitrogen immediately and stored at À 80 � C until total RNA extraction.

Cell lines and cell cultures
The CC cell lines including SiHa, CaSki, and HeLa, and a normal cervical epithelial cell line H8 were purchased from the Department of Cell Biology, the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences (Beijing, China).These cells were cultured in RPMI 1640 medium (Gibco-BRL, Grand Island, CA, USA) with 10% foetal bovine serum (FBS; Gibco-BRL), 100 lg/mL streptomycin, and 100 U/mL penicillin at 37 � C with a humidified 5% CO 2 environment.

RNA extraction and ribonuclease R (RNase R) treatment
Total RNAs from serum and cells were extracted using the serum RNA extraction kit (Takara Bio, Shiga, Japan) and the TRIzol reagent (Invitrogen, Carlsbad, CA, USA), respectively, following the manufacturer's protocol.The concentration and purity of RNA samples were detected by a spectrophotometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA), and the OD260/OD280 ratio between 1.8 and 2.0 indicated good purity.RNase R (Epicentre Technologies, Madison, WI, USA) was employed to degrade linear RNAs.Briefly, 2 lg of total RNA was mixed with 3 U/lg of RNase R, and then incubated at 37 � C for 15 min.

Statistical analysis
Statistical analysis was performed using SPSS Statistics Version 27.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism Version 5.0 (GraphPad Software, Inc., La Jolla, CA, USA).Each experiment was performed in triplicate and the results were presented as the mean ± standard deviation (SD).Levels of circ_0000285 were compared between groups via Student's t-test or one-way ANOVA.Spearman correlation analysis was used to analyse the correlations between circ_0000285 expression and clinical characteristics of CC patients.Receiver operating characteristic (ROC) curves and the area under the curve (AUC) were used to assess the diagnostic values of circ_0000285, CA125, and SCC-Ag for CC and ESCC.The binary logistic regression analysis was used to evaluate the combination of circ_0000285, CA125, SCC-Ag, and HPV as a diagnostic biomarker for CC and ESCC.The cut-off values were calculated using Youden's index.All p values were twosided, and p < 0.05 was considered statistically significant.

Comparison of baseline characteristics between CC patients and healthy volunteers
The baseline data of CC patients (n ¼ 152) and healthy volunteers (n ¼ 116) were shown in Table 1.No significant differences were observed in age, body mass index, smoking, and drinking in the two groups.However, the CA125 and SCC-Ag levels and HPV-positive rates were significantly higher in CC patients than in healthy controls (P < 0.01).

Expression levels of circ_0000285 in serum and cervical exfoliated cells from CC patients
The levels of circ_0000285 in serum from 116 healthy volunteers, 65 ESCC patients, and 87 LACC patients preoperative were examined by qRT-PCR, and found that circ_0000285 expression was upregulated in serum and cervical exfoliated cells from CC patients than that in healthy volunteers (Figure 1(A, B), p < 0.05).And high levels of circ_0000285 were observed in serum and cervical exfoliated cells from LACC patients compared to that of ESCC patients (Figure 1(C, D), p < 0.05).Similarly, the expression levels of circ_0000285 were upregulated with 4.94, 2.53 and 2.81 fold changes in the SiHa, CaSki, and HeLa cells, respectively, compared to the H8 cells (Figure 1(E), p < 0.01).Besides, circ_0000285, but not HIPK3 mRNA, was resistant to RNase R treatment, further confirming the circular structure of circ_0000285 (Figure 1(F), p < 0.001).Taken together, these results herein demonstrated that the expression of circ_0000285 is upregulated in serum and cervical exfoliated cells from CC patients.

Correlations between circ_0000285 expression and clinical characteristics of CC patients
The expression levels of circ_0000285 were higher in serum and cervical exfoliated cells from HPV positive CC patients than that in HPV negative CC patients (Supplementary Figure 1(A,B), p < 0.05).Similarly, the expression levels of circ_0000285 were upregulated in CC patients with high SCC-Ag levels (� 4 lg/L) compared to that of CC patients with low SCC-Ag levels (< 4 lg/L) (Supplementary Figure 1  The expression levels of circ_0000285 in the serum of CC patients preoperative and 90 days after surgery were shown in Figure 2(A).Compared to preoperative, the expression levels of serum circ_0000285 were significantly decreased in CC patients after surgery (Figure 2(B), p < 0.001).Interestingly, as shown in Figure 2(C), CC patients with high serum circ_ 0000285 expression after surgery had a higher risk of tumour relapse than those CC patients with low serum circ_0000285 expression (p < 0.05).These results revealed that high circ_ 0000285 expression correlates with tumour relapse in CC patients after surgery.

Evaluation of circ_0000285 in serum and cervical exfoliated cells as a novel diagnostic biomarker for CC
To investigate the diagnostic values of the combination of circ_0000285 in serum and cervical exfoliated cells in distinguishing CC patients from healthy volunteers, ROC curves were plotted and the results showed that the AUC was 0.855 (95% CI: 0.812-0.898),indicating a high diagnostic value of circ_0000285.The cut-off value used for circ_0000285 was 9.52, and the sensitivity and specificity were 0.644 and 0.879,     serum and cervical exfoliated cells could serve as an effective diagnostic biomarker in CC.
The integration analysis showed that circ_0000285 had a better diagnostic value than CA125 and SCC-Ag alone, and the combination of circ_0000285, CA125, SCC-A, and HPV showed the highest diagnostic value (AUC: 0.894, 95% CI: 0.836-0.951,sensitivity: 0.892, specificity: 0.885) (p < 0.001).The above results demonstrated the possible use of circ_0000285 in serum and cervical exfoliated cells as a diagnostic biomarker in ESCC.

Discussion
Improvement on the basis of early diagnosis is still the main approach for achieving long-term cancer-free survival.
Although several protein biomarkers of malignant tumours are currently used in the clinical setting, these biomarkers have limitations in diagnosis of ESCC.In recent years, circRNAs have caused great attention and have been developing rapidly in clinical diagnosis of human diseases, especially in malignant tumours (Sole et al. 2021).Because of easy extraction, high stability, and high specificity than proteins, circRNAs are superior to current tumour biomarkers in many aspects and provide a new perspective in the diagnosis of cancers.For example, circ_MBOAT2 is reported overexpressed in tissues and serum samples of colorectal cancer, this circRNA may be a novel potential biomarker of colorectal cancer (Tang et al. 2020).Fan et al. (2019) reported that circ_ MAN1A2, which is upregulated in nasopharyngeal carcinoma (NPC), oral cancer, thyroid cancer, ovarian cancer, and lung cancer, could be a serum biomarker for these malignant tumours.Guo et al. (2022) suggested that serum hsa_circ_ 0079480 is a novel prognostic marker for acute myeloid leukaemia.In the present study, we found that circ_0000285 expression in serum and cervical exfoliated cells from CC patients preoperative was significantly increased as compared to those of the healthy volunteers.We also showed that circ_ 0000285 might be used as a novel, non-invasive biomarker for ESCC in clinical practice.The circ_0000285, derived from exonic back-splicing of HIPK3 gene, which is encoded on chr11: 33362513-33363232 location, has been identified as an oncogenic driver of CC capable of promoting tumour development through the miR-197-3p-ELK1 axis (Zhang and Zhang 2020).Unsurprisingly, the circular structure of circ_0000285 was verified by RNase R treatment.The present study confirmed that the upregulated expression of circ_0000285 in serum and cervical exfoliated cells from CC patients, which was consistent with circ_ 0000285 expression in cancer samples of CC reported by Chen et al. (2019) and Zhang et al. (2022).Additionally, highly expressed circ_0000285 was also reported in other malignant tumours, including thyroid cancer (Gao et al. 2021), osteosarcoma (Long et al. 2020;Zhang et al. 2020), hepatocellular carcinoma (Huang et al. 2022), and glioma (Liu et al. 2022).Similarly, we found circ_0000285 expression was significantly increased in the SiHa, CaSki and HeLa cells compared with the H8 cells.Based on the above results, we believed that highly expressed circ_0000285 in serum and cervical exfoliated cells from CC patients is credible.
The circ_0000285 is previously reported to serve as a potential biomarker in solid cancer.In NPC, Shuai et al. (2018) demonstrated that circ_0000285 is a novel biomarker for NPC prognosis and is involved in radiosensitivity.In bladder cancer, circ_0000285 serves as a novel prognostic biomarker with involvement in chemosensitivity (Chi et al. 2019).In gastric cancer, circ_0000285 acts as a potential therapeutic biomarker by regulating tumour progression through miR-1278-FN1 axis (Wang et al. 2022).Here, our further analysis showed that circ_0000285 expression was increased in preoperative LACC patients than that in ESCC patients.Serum circ_0000285 expression was significantly decreased in CC patients after surgery.And the CC patients after surgery with high serum circ_0000285 expression were more prone to have a tumour relapse.In addition, we found that high serum circ_0000285 expression was positively correlated with SCC-Ag level and HPV positive rate.All these findings provided the theoretical clues to support the possibility of circ_ 0000285 as an adjuvant diagnostic marker for CC.
In order to identify the diagnostic values of the combination of circ_0000285 in serum and cervical exfoliated cells for CC, ROC curves analysis was performed.We found that the AUC of circ_0000285 was superior to CA125 and SCC.More importantly, the combination detection of circ_ 0000285, CA125, SCC, and HPV could greatly improve the diagnostic efficiency of CC.Besides, circ_0000285 had significant diagnostic value in distinguishing ESCC patients from LACC patients.The strength of this study is the elucidation of the high potential of circ_0000285 in serum and cervical exfoliated cells as a diagnostic biomarker for ESCC.
To date, increasing data supported the possibility to use several specific biomarkers to distinguish LACC and ESCC.For example, peritoneal HPV positivity (PHP) can be used for evaluation of initial peritoneal disease development in LACC (Bizzarri et al. 2021).The cell-free HPV DNA (ctHPV DNA) is a promising biomarker in LACC (Sivars et al. 2022).Based on the above evidence, it would be valuable to analyse whether the combination of circ_0000285, PHP and ctHPV DNA is better than alone for the diagnosis of ESCC.
There are several limitations to our study: 1) CC patients cohort is relatively small; 2) the heterogeneity of included patients; 3) Since the follow-up time was not long enough, the relationships between circ_0000285 expression and overall survival and disease-free survival of CC patients are unclear; 4) As circ_0000285 expression is correlated with HPV positive rate, whether circ_0000285 is involved in HPV infection-mediated cervical carcinogenesis remains to be confirmed.
Based on the above data, it was concluded that circ_ 0000285 expression is upregulated in serum and cervical exfoliated cells of CC patients, and correlates with SCC-Ag level, HPV positive rate, and tumour relapse.Our data preliminary demonstrated that highly expressed circ_0000285 may be a superior biomarker for the diagnosis of ESCC than CA125 and SCC-Ag.Further research are needed to focus on clinical large-scale cohort, prospective, multi-sample, and multi-center studies to further confirm and support the findings in our study.

Figure 1 .
Figure 1.The circ_0000285 is highly expressed in serum and cervical exfoliated cells of CC patients.(A, B) qRT-PCR analysis of the expression levels of circ_0000285 in serum and cervical exfoliated cells of healthy volunteers (n ¼ 116) and 152 CC patients (2 À DDCt method).(C, D) Analysis of the expression levels of circ_0000285 in serum and cervical exfoliated cells of ESCC patients (n ¼ 65) and LACC patients (n ¼ 87) using qRT-PCR.(E) The expression levels of circ_0000285 were significantly higher in the SiHa, CaSki and HeLa cells compared to the normal cervical epithelial cell line H8.(F) The expression levels of circ_0000285 and HIPK3 mRNA were analysed in SiHa cells after treatment with RNase R. CC: cervical cancer; qRT-PCR: quantitative real-time polymerase chain reaction; ESCC: early-stage CC; LACC: locally advanced CC; HIPK3: homeodomain interacting protein kinase 3; RNase R: Ribonuclease R; SD: standard deviation.Data were presented as mean ± SD, � p < 0.05, �� p < 0.01, ��� p < 0.001, ns p > 0.05.

Figure 2 .
Figure 2. The expression of circ_0000285 is downregulated in serum from CC patients after surgery.(A) qRT-PCR analysis of circ_0000285 expression in serum from CC patients preoperative and 90 days after surgery.(B) Decreased serum circ_0000285 expression was observed in CC patients after surgery.(C) The CC patients after surgery with high serum circ_0000285 expression was more prone to have a tumour relapse.Data were presented as mean ± SD, � P < 0.05, ��� P < 0.001.

Table 2 .
Correlations between the expression levels of circ_0000285 in serum and cervical exfoliated cells and clinical characteristics of CC patients.