Anti-Silencing Function 1B Overexpression Affects Prognosis and Promotes Invasion in Cervical Carcinoma via Activation of Wnt/β-Catenin Signaling Pathway

Background: Anti-silencing function 1B histone chaperone (ASF1B) has been identied to compensate for growth defects and sensitivity to replication stress. Recent studies demonstrated that ASF1B plays an important role in the tumorigenesis of human cancer. However, its relationship with clinical outcome of cervical carcinoma is unknown. Therefore, we aimed to investigate the expression pattern and clinical signicance of ASF1B in cervical cancer and its role in regulating invasion and metastasis of cervical cancer cell lines. Methods: Expression of ASF1B was analyzed in eight cervical cancer cell lines, and eight pairs of cervical cancer samples and the adjacent normal specimens, using real-time PCR and western blotting. Immunohistochemistry was used to analyze ASF1B expression in paran-embedded tissues from 147 cervical cancer patients. The association between ASF1B expression levels, clinicopathological parameters, and prognosis was analyzed statistically. Moreover, the biological function and potential mechanism of ASF1B in migration and invasion of cervical cancer cells were investigated by in vitro experiments and Western blotting. Results: ASF1B mRNA and protein levels were overexpressed in cervical cancer cell lines and tissues compared with normal cells and adjacent normal cervical specimens. In paran-embedded cervical carcinoma tissues, upregulation of ASF1B protein was identied in 86 (58.5%) of 147 cases, and was remarkably associated with pelvic lymph node metastasis, lymphovascular space involvement, property of surgical margin, FIGO stage, sqaumous cell carcinoma antigen and poor survival. Cox analysis demonstrated that ASF1B expression was an independent risk predictor for survival in patients with cervical carcinoma. Additionally, down-regulation of ASF1B signicantly inhibited invasion and migration of cervical cancer cells. Moreover, it was demonstrated that ASF1B regulated the Wnt/β-catenin pathway in cervical carcinoma. Conclusions: This study suggested that ASF1B might serve as a important prognostic biomarker and therapeutic target for patients with cervical carcinoma.


Introduction
Cervical cancer is the third most common type of malignant gynecological tumor worldwide, ranking as the fourth main cause of cancer-associated deaths among women [1]. However, in view of the vast population base, it is reported that China accounts for 11.7 % (62,000 new cases) of the world's annual cervical cancer morbidity and 11.3 % (30,000 deaths) of annual cervical carcinoma mortality worldwide [2]. Despite current efforts and advances in early detection, treatment strategies, and research of cervical carcinoma, which have effectively lower the mortality rate of this malignancy, cervical carcinoma remains a prominent public health and economic burden in China. It is reported that patients with higher stage of International Federation of Gynecology and Obstetrics (FIGO), lymph node metastasis (LNM), lymphovascular space involvement or higher differentiation grade have poorer clinical outcomes [3]. In the clinic, the squamous cell carcinoma antigen (SCC-Ag), an important serum biomarker of cervical carcinoma, has been used to determine LNM, prognosis, and recurrence in cervical cancer patients [4].
Moreover, a variety of novel oncogenes have been identi ed as signi cant biomarkers that are associated with unsatisfactory clinical outcome of cervical neoplasm [5][6][7]. However, they are not su ciently powerful to predict clinical outcome, and their clinical applications are limited because of poor sensitivity and speci city. Hence, identifying new noninvasive, reliable, and effective biomarkers for the prognosis and treatment of cervical cancer, as well to improve our understanding of the mechanism of this disease, are urgently required.
ASF1B (anti-silencing function 1B histone chaperone), located on human chromosome 19 at 19p13.12, encodes a member of the H3/H4 family of histone chaperone proteins, which served an important role in histone dynamics during replication, transcription, and DNA repair [8]. Messiaen et al. indicated that ASF1B expression was speci cally found in the germ cell lineage and ASF1B peak expression was signi cantly associated with meiosis [9]. Upregulation of ASF1B resulted in different transcriptional signatures consistent with enhanced cellular proliferation and decreased cellular apoptosis [10]. Previous studies indicated that ASF1B exerts important roles in the development of human cancer. Corpet et al. [11] showed that ASF1B is a prognostic predictor and proliferation indicator in early stage breast cancer.
ASF1B expression is important for cellular proliferation, possibly during and after S phase, which is of great signi cance in progression of neoplasm. A recent study revealed that ASF1B, known as one of several chromatin regulators, is associated with local and metastatic relapse, and poor patient outcome in breast carcinoma [12].
In this study, we investigated the expression pattern of ASF1B in cancer of uterine cervix. The present study detected expression levels of ASF1B mRNA and protein in cervical cancer cell lines and surgical tissues of patients with cervical carcinoma. Correlations between ASF1B protein expression, and the clinicopathological risk factors and patient clinical outcomes were further analyzed. Furthermore, the function of ASF1B in uterine cervical cancer cell lines and its possible mechanism were investigated.

Cell lines and treatment
We purchased eight cell lines of cervical cancer (HeLa, C33A, SiHa, HeLa229, MS751, ME180, HCC94, and CasKi) from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China), which were cultured in Roswell Park Memorial Institute RPMI-1640 medium (GIBCO, Gaithersburg, MD, USA).

Samples and patient cohorts
The present study was carried out based on a cohort of 147 para n-embedded, surgical tissues of earlystage cervical carcinoma. All cases were clinically and histopathologically diagnosed with primary cervical carcinoma at the Sun Yat-sen University Cancer Center between 2006 to 2011. The FIGO (2009) standard was utilzed for peer-review of the clinical diagnoses and grading of the cases. World Health Organization (WHO) criteria was used for diagnoses of histological types and grades. For study purposes, the approval from the Medical Ethics Committee of the Sun Yat-sen University Cancer Center were obtained.
In this cohort, the 147 patients were diagnosed with IB1-IIA2 stage cervical carcinoma and all received a radical hysterectomy and lymphadenectomy. None of the patients had received preoperative radiotherapy, chemotherapy, or molecular-targeted therapy. Patients were followed up regularly in the clinic. We excluded patients who were lost of follow-up or loss of follow-up for short term. In the present study, according to the guidelines, sufferers with high-risk factors were arranged to receive radiotherapy and/or chemotherapy after surgical operations. The high-risk indicators comprise of PLNM, large tumor size (> 4 cm), deep stromal invasion, positive surgical margin, positive parametrial involvement, high differentiation grade, and positive lymphovascular space involvement. No patients received immunotherapy after surgical treatment.
Eight collected uterine cervical neoplasm samples and matched tumor-adjacent normal cervical tissues were frozen at − 80 °C and laid in liquid nitrogen prior to experiments. In addition, demographic and clinicopathological information for the 147 participants mentioned above were retrieved from impatient medical records and are summed up in Table 1. The median age of the cohorts was 46 years so that we used it as an age gap to divide the cases into two groups. In accordance to FIGO criteria, 61 patients were labeled as stage IB1; 24 as stage IB2; 42 as stage IIA1; and 20 as stage IIA2. The Total RNA from cell or tissue was extracted utilizing the Tri-Reagent (Sigma-Aldrich; Merck KGaA) on the basis of the manufacturer's protocol. The RNA quality was detected by An Agilent Bioanalyzer 2100. First-strand cDNA synthesis was carried out utilizing a Prime Script Reverse Transcriptase reagent Kit (Takara, Japan). Complementary DNA (cDNA) was ampli ed and quanti ed as follows. An initial ampli cation reaction utilizing speci c primers of ASF1B was performed with a denaturation step at 95°C for 10 min, followed by 30 cycles at 95°C for 10 sec, primer annealing/extension at 60°C for 1 min. A nal extension at 72°C for 5 min was carried out before the reaction was stopped and stored at 4°C. The PCR primers sequences of ASF1B are as follows: forward 5′-TGGACAGGAGTTCATCCGAGTG-3′ and reverse 5′-GTCCCAGTTGATATGGAAGCGG-3′. The 2−ÄÄCt (cycle threshold) method were used, where the expression of GADPH was regarded as an endogenous standard to normalize the relative changes in ASF1B expression [5]. Primers for GADPH are as follows: forward 5'-AGAAGGCTGGGGCTCATTTGC-3' and reverse 5'-ACAGTCTTCTGGGTGGCAGTG-3'. All experiments were carried out in biological triplicate.

Western blotting analysis
Protein was extracted using RIPA buffer (BiYunTian, Shanghai, China ) according to the manufacturer's protocol. Western blotting was carried out utilizing standard procedures according to our previous publication [3]. The protein samples (30 µg) were separated on 9% sodium dodecyl sulfate polyacrylamide gels and transferred onto polyvinylidene uoride microporous membranes (Millipore, Billerica, MA, USA ). After blocking with 5% non-fat milk for 2 h, the membranes were incubated with primary antibodies against ASF1B ( Immunohistochemical staining and assessment Immunohistochemistry (IHC) analysis was performed according to standard protocols, as previously mentioned [3] with anti-ASF1B antibodies (No. 11011-1-AP, Proteintech, USA). Para n-embedded specimens were prepared for IHC experiments from 147 patients with early-stage cervical carcinoma. Brie y, the 4-µm thickness para n-embedded sections were cut and baked at 60 °C for 30 min. The slides were then depara nized for 10 min in xylene (twice), and rehydrated through graduated alcohols to running distilled water to rinse. Submerging the sections into EDTA antigenic retrieval buffer and then microwaving were used for antigen retrieval. The tissue sections were blocked with 3 % hydrogen peroxide in methanol to quench the endogenous peroxidase activity, and then, the slides were incubated with 1 % bovine serum albumin for blocking nonspeci c binding. The samples were rinsed with phosphate-buffered saline (PBS, PH 7.4). Next, the tissues were incubated with the primary antibodies against ASF1B ( 1:200 ) at 4 °C overnight. After washing, tissue sections were incubated with biotinylated anti-rabbit secondary antibody (1:1000; Abcam) at room temperature, followed by incubation with streptavidin-horseradish peroxidase complex (Abcam). The samples were then submerged in 3,3diaminobenzidine (DAB) and counterstained with 10% Mayer's hematoxylin, dehydrated, and mounted.
Each section was evaluated independently by two pathologists. Sections were viewed under a bright eld microscope. For ASF1B protein evaluation, the average staining intensity (SI) (0 = none; 1 = weak; 2 = modest and 3 = strong) of immunoreactive tumor cells were assessed. The percentage of positive stained tumor cells was given score as follows: 1 (1-25% ); 2 (26-50% ); 3 (51-75% ), or 4 (> 75% ). The staining index was calculated as the product of combining sensitivity and speci city. The log-rank test statistical analysis was applied to select the optimum cutoff score. Low expression of ASF1B protein was classi ed as 0 to 4 scores; high expression of ASF1B protein was classi ed as > 4 scores.

Cell invasion assay
Migration of cells was assessed by cell invasion assay. Transwell insert chambers (Corning Incorporated, Corning, NY, USA ) with 8-µm Matrigel-coated (BD Biosciences, USA) pore polycarbonate lters were used to test it. Cells plated in the absence or presence of siRNA at the density of 2×10⁴ cells/well were incubated into the upper chamber with serum-free medium and allowed to invade in the lower chamber with 10% FBS medium as a chemotaxis agent. After 48h incubation, migratory cells on the lower surface of the membrane were stained with 0.5% crystal violet, examined by phase contrast microscopy(×200 magni cation), photographed and counted. We performed all the experiments in triplicate independently.
Wound-healing assay Cells(1 × 10⁴ per well) were cultivated in a six-well plate separately. Untransfected cells and cells transfected wtih ASF1B siRNA were incubated in a quiescent medium overnight. Next, in con uent monolayers, a sterile 10-µl pipette tip was applied to gently and slowly scratched across the center of the well. The cells were washed with 1× PBS and replaced with fresh serum-free medium. After wounding by Olympus IX73 microscope, the microscopic pictures were captured at 0 and 24h subsequently. Values were means ± standard deviation (SD) from at least three independent assays.

Statistical analysis
Statistical analyses were performed utilizing SPSS 17.0 statistical software package (SPSS Inc, Chicago, IL, USA). The association between ASF1B protein expression and histomorphological factors was accessed by Pearson' chi-squared test and Fisher's exact test. Spearman's rank correlation coe cients were applied to evaluate the bivariate correlations between the variable factors. The overall survival (OS) and the disease-free survival (DFS) in correlation with ASF1B protein expression were calculated using Kaplan-Meier analysis and compared using the log-rank test. Cox regression analyses(univariate and multivariate survival analyses) were performed to investigate survival data. P values less than 0.05 were considered statistically signi cant in all cases.

Results
Expression of ASF1B is elevated in human cervical cancer Expression pattern of ASF1B mRNA and protein were examined in cervical cancer cell lines and tissues using qRT-PCR and analysis of western blotting. Compared with normal cervical samples, the ASF1B mRNA and protein were signi cantly overexpressed in cell lines of cervical cancer ( Fig. 1, P < 0.05). Consistent with the results in cell lines, ASF1B transcriptional and translational expressions were signi cantly higher in the eight cervical neoplasm specimens than their expressions in the matched adjacent normal samples ( Fig. 2, P < 0.05). Next, the ASF1B expression in 147 lesions was further investigated using IHC. Consistent with the above results, IHC showed elevated levels of the ASF1B protein expression in early-stage utertine cervical carcinoma specimens (Fig. 2C).
Overexpression of ASF1B protein correlates with the clinicopathologic characteristics of early-stage cervical carcinoma To determine the relationship between ASF1B levels and clinical pathological parameters, and whether ASF1B could act as a new important prognostic biomarker, we determined the ASF1B levels in a cohort of 147 para n-embedded existing cervical neoplasm samples by IHC. The cervical carcinoma specimens for IHC included FIGO stage IB1 (61 cases, 41.5%), stage IB2 (24 cases, 16.3%), stage IIA1 (42 cases, 28.6%), and stage IIA2 (20 cases, 13.9%) disease. IHC analysis showed that ASF1B protein expression level was evaluated as negative or weakly positive in 61/147cases (41.5%), and positive in 86/147 (58.5%) cases. Positivity for ASF1B staining was mostly localized in the nucleus and rarely in epithelial cells cytoplasm (Fig. 3A). The mean response intensity of all positive cervical cancer cells in the eld of vision was explored utilizing Mean optical density (MOD). The quantitative analysis demonstrated that the MOD of ASF1B protein were signi cantly increased in the subgroups with lymphovascular space involvement compared with that in the subgroups without lymphovascular space involvement (Fig. 3B, P < 0.05). The MOD values of ASF1B protein were lower in the subgroups without PLNM than in the subgroups with PLNM ( Fig. 3C, P < 0.05). Furthermore, the correlation between the protein expression of ASF1B and pathological parameters was explored utilizing the Fisher's exact test and Chi-squared test. As summarized in Table 2, we didn't nd any obvious association of ASF1B protein levels with age, histological type, tumor size, myometrium invasion, in ltration of the parauterine organ, tumor recurrence, or differentiation grade (P > 0.05). However, ASF1B protein overexpression was signi cantly associated with SCC-Ag (P = 0.037), FIGO stage (P = 0.010), property of surgical margin (P = 0.036), lymphovascular space involvement(P < 0.001), PLNM (P = 0.022), and even vital status (P < 0.001). The above relationships were further statistically analysed by Spearman's rank correlation test, which is summarized in Table 3.  To assess the association of ASF1B with survival in cervical neoplasm, we evaluated the relationship between the expression levels of ASF1B protein and clinical outcome in cohorts with early-stage cervical carcinoma. Notably, the survival curves showed that patients with ASF1B protein overexpression exhibited poorer OS and DFS compared with those with low expression of ASF1B protein in early-stage cervical cancer (log-rank, P < 0.001; Fig. 4A and 4B). In the low-ASF1B group, the cumulative rates of OS and DFS were 96.6% and 90.9% respectively while they were 57.7% and 78.0% respectively in the high-ASF1B group.
Additionally, the survival values of ASF1B protein expression pattern were calculated in different subgroups of cervical cancer cohorts, which were classi ed according to SCC-Ag, PLNM, age, tumor size, FIGO stage, properties of the surgical margin, differentiation grade, deep stromal invasion, lymphovascular space involvement, chemotherapy, radiotherapy, and parauterine organ in ltration.
We subsequently detect whether ASF1B protein could serve as a novel prognostic factor using cox regression analysis. As detailed in Table 4, univariate analysis showed that ASF1B protein expression, PLNM, lymphovascular space involvement, and recurrence were strongly correlated with OS (P < 0.05).

Silencing of ASF1B attenuates migration and invasion in CasKi and Siha cells
Our result found that ASF1B protein is signi cantly correlated to lymph node metastasis. As we know, cell migration and invasion are strongly associated with tumor metastasis. Thus, we analysed gene expression array data publicly available online, which was used in Gene Set Enrichment Analysis (GSEA) for ASF1B. The result showed that upregulation of ASF1B was correlated with tumor metastasis (Fig. 7A).
Next, we carried out a transwell migration assay and a wound healing assay. In transwell assay, the results showed that silencing ASF1B obviously reduced the number of cells on membrane lters (Fig. 7B).
Consistently, the results of wound healing assays found that the effect of ASF1B inhibition on the invasiveness of CasKi and Siha cells was remarkably hampered (Fig. 7C). Overall, these results suggested that inhibition of ASF1B weakened the invasion and migration properties of cervical cancer cells.
Silencing of ASF1B inhibited the Wnt/β-catenin signaling pathway in cerivcal carcinoma We used GSEA to further detect the potential mechanism by which ASF1B promotes the invasion and migration capacities of cervical carcinoma cells. The result demonstrated that upregulation of ASF1B was correlated with Wnt/β-catenin signaling pathway (Fig. 8A). Since Wnt/β-catenin signalling pathway is regarded as a key regulator of CSC self-renewal, we hypothesized that ASF1B may activate Wnt/βcatenin signalling pathway and therefore promote stem cell-like capacities and tumourigenic potential in cervical neoplasm. We detected the effect of ASF1B on the subcellular localization of β-catenin to validate this hypothesis. Interestingly, repression of ASF1B expression hampered, the translocation of βcatenin into the nucleus, as demonstrated by Western blotting results of nuclear and cytoplasmic cellular components (Fig. 8C). Next, we identi ed that the expression pattern of Wnt/β-catenin signaling pathwayrelated genes, including c-Myc and cyclin D1. The results found that the expression levels of c-Myc and cyclin D1 proteins were strongly decreased in ASF1B silencing cells (Fig. 8B). All the data mentioned above indicated that ASF1B may regulate the Wnt/β-catenin signaling pathway in cervical carcinoma.

Discussion
The current study showed that compared with that in matched adjacent non-cancerous cervical specimens, the ASF1B expression was remarkably overexpressed in human cervical carcinoma lesions at both the transcription and translation levels. High ASF1B protein levels were associated with clinical characteristics such as FIGO stage, SCC-Ag, PLNM, lymphovascular space involvement, property of surgical margin, and even vital status. The results showed that measurement of the ASF1B protein level is a promising independent tool to test or predict survival of the sufferers with early-stage cervical carcinoma. Furthermore, in vitro assays indicated that inhibition of ASF1B expression hampered the invasiveness and migration of cervical cancer cells through regulating Wnt/β-catenin signaling pathway.
Our results revealed a mechanism that enhances tumor metastasis in cervical neoplasm, which implied that ASF1B had potential to be an important therapeutic target for cervical carcinoma treatment.
In previous studies, ASF1B was proven to have an important role in tumors, which is consistent with ndings from our study. ASF1B exists as a distinct isoform of ASF1 in humans. ASF1, originally found in budding yeast because of its effect on heterochromatin silencing, is the most conserved histone H3/H4 chaperone protein from yeast to humans. The most important ASF1-dependent acetylation target sites in mammals are H3K56 and H3K56 acetylation, which are culpably involved in genome stability, DNA replication, stem cell pluripotency, and cancers [13][14][15][16]. Seol et al. [17] found that H3K56 acetylation increased with increasing levels of ASF1B in Hela cells, which makes ASF1B an fascinating therapeutic target for cancer treatment. It has been demonstrated that ASF1B served as a novel survival-related biomarker that correlates with poor tumor grade with relevance to the clinical outcome in breast cancer [11][12]. Moreover, ASF1B is an important prognostic factor for tumor metastasis and exerts a critical role in cell proliferation. ASF1B was highlighted in the prominent biology, which was related to the response of colon carcinoma cells to bortezomib. ASF1B played an important role in proteasome depression-induced cell apoptosis, exactly polyamine dysregulation, mTOR ablation, and DNA damage signaling enhancement [18]. Co-exhaustion of the histone chaperones ASF1A and ASF1B yielded all typical biomarks of alternative lengthening of telomeres (ALT) in both primary and tumor cells, which is positively regulated by RAD17 checkpoint clamp loader component (RAD17) and Bloom syndrome RecQ like helicase (BLM), and negatively regulated by exonuclease 1 (EXO1) and DNA Replication Helicase/Nuclease 2 (DNA2) [19]. In addition, ASF1B is targeted by microRNA miR-214 in myeloma.
These studies indicated that ASF1B may be directly or indirectly involved in the progression of various tumors through different molecular mechanisms. In our cohort, ASF1B was found up-regulated at both the transcriptional and translational levels in cervical carcinoma. We showed the relationship between ASF1B expression and some clinical high risk factors, which indicated that ASF1B is an important predictor that correlates with tumor development and progression in cervical carcinoma. Collectively, our results suggested that ASF1B overexpression is a new independent indicator for poor survival in human malignancies. Wnt/β-catenin signaling pathway exerts a regulatory role in a series of cellular events, including cell invasion, proliferation and differentiation via regulating the capacity of the multifunctional β-catenin protein [20]. It is reported that Wnt/β-catenin signaling is often aberrantly activated during the carcinogenesis and aggressiveness of cervical malignancy [21]. Wnt pathway was previously reported to exert a great impact on regulation of cell migration and appears to be an important step during cervical carcinogenesis [22]. Nevertheless, little is known about the correlation between ASF1B and Wnt/β-catenin signaling in cervical cancer metastasis. Our results indicated that inhibition of ASF1B may repress invasion and migration via Wnt/β-catenin signaling in uterine cervical neoplasm. These data mentioned above have provided new insights into the tumor development and progression of cervical cancer.
However, the precise mechanisms that link ASF1B to the metastatic process in cervical cancer are unclear and require further research.
Treatment strategies are strongly correlated with prognosis in cervical cancer patients. Various treatment strategies cause different types and degrees of side effects in patients. Clinically, the current standard treatment for early-stage cervical carcinoma remains radical hysterectomy, plus lymphadenectomy (RH) or concurrent chemoradiotherapy (CCRT). According to the guidelines for the treatment of cervical carcinoma, such as those of the National Comprehensive Cancer Network [23], National Cancer Institute for patients with early-stage cervical neoplasm [26]. Patients who receive CCRT instead of unnecessary surgical treatment can avoid surgery-related side effects, including intra-and postoperative bleeding, in ammation, deep vein thrombosis, pelvic lymphocytes, lymphedema (tissue swelling), and elephantiasis (tissue/skin thickening). Furthermore, lymphovascular space involvement is an independent risk factor that is remarkably associated with poor clinical outcome in cervical carcinoma [27]. Thus, it is important to identify patients with positive lymphovascular space involvement. Deng et al. [28] found that Stomatin-like protein 2 (SLP-2) may be a progressive gene in the development of cervical cancer and correlates with lymphovascular space involvement. It is reported that WNT2 overexpression was correlated with lymphovascular space involvement in cervical cancer [29]. Moreover, overexpression of the kinesin family member 20A (KIF20A) protein was demonstrated to be strongly associated with lymphovascular space involvement in carcinoma of the uterine cervix [30]. However, currently, few suitable tumor markers can be applied to predict lymphovascular space involvement in the clinical setting. Here, our data revealed that upregulation of ASF1B is signi cantly associated with positive lymphovascular space involvement in cervical cancer. Additionally, a relationship between OS and ASF1B protein overexpression was detected in the "without lymphovascular space involvement" and "with lymphovascular space involvement" subgroups. Finally, a signi cant correlation between poorer OS and ASF1B upregulation was observed in the subgroup of "without lymphovascular space involvement", which indicated that ASF1B might serve as a potential marker for clinical outcome prediction of patients with cervical cancer without lymphovascular space involvement. Nevertheless, Cox regression analysis determined that lymphovascular space involvement is not an independent prognostic biomarker of cervical cancer. One explanation may be that there were insu cient cases of lymphovascular space involvement among the recruited patients. Thus, relevant research extended to a larger cohort of patients diagnosed with lymphovascular space involvement is needed. More studies should be planned and performed to explore the related molecular mechanisms of the relationship between ASF1B expression and lymphovascular space involvement in cervical carcinoma.
Pelvic lymph node metastasis (PLNM) exerts a pitoval and direct role in prediction of the survival of cervical cancer patients, additionally determining the optimal therapeutic arrangements [31]. For patients with PLNM, CCRT is needed, which in retrospect, would cause the initial surgical procedure unnecessary.
Currently, no precise molecular marker has been found to identify patients with PLNM to avoid improper treatment and access to better clinical decision making. Thus, there is an urgent need to identify an ideal preoperative marker to predict PLNM in cervical carcinoma. The current study showed that aberrant ASF1B protein expression was largely correlated with PLNM in cervical carcinoma. In univariate analysis, the result suggested that PLNM might be a prognostic predictor in patients with early-stage cervical carcinoma, which is similar to previous studies [32][33]. Notably, ASF1B protein overexpression was remarkably associated with worse postoperative OS of patients with cervical carcinoma in subgroups with or without PLNM. This indicated that different levels of ASF1B protein expression can help clinics to identify patients with poorer clinical outcome, which will lead to more aggressive treatment for these sufferers to decrease the tumor-related deaths. Nevertheless, no signi cant correlation was determined between ASF1B protein expression and PLNM in cervical cancer. In future work, extensive researches with a larger-scale sample are necessary to understand the role and detailed mechanisms of ASF1B protein expression's effects on PLNM in cervical carcinoma.
In conclusion, this study indicated that ASF1B is overexpressed in cervical carcinoma and elevated ASF1B expression is an independent prognostic predictor which signi cantly associated with SCC-Ag,