The Prognostic Value of Preoperative Fibrinogen in Patients with Radical Cholecystectomy

Background and Aims . There is currently a lack of suitable hematological markers as a complement to pathological factors in predicting the prognosis of patients with gallbladder cancer. The study aimed to investigate the clinical value of preoperative ﬁ brinogen levels in assessing the prognosis of patients with gallbladder cancer after radical surgery. Methods . The study retrospectively analyzed 260 gallbladder cancer patients who underwent radical resection. Time-dependent receiver operating characteristic (ROC) curves were used to calculate the optimal cut-o ﬀ values of carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9), and ﬁ brinogen. Besides, univariate and multivariate analyses of disease-free survival (DFS) and overall survival (OS) were performed to determine independent markers in peripheral blood. Then, subgroup analyses of ﬁ brinogen in di ﬀ erent degrees of di ﬀ erentiation, age, gender, and BMI were performed by log-rank test. Result . The cut-o ﬀ values of ﬁ brinogen, CEA, and CA19-9 were set at 2.97g/L, 2.17ng/mL, and 41.1U/mL, respectively. The results showed that the preoperative ﬁ brinogen level was associated with tumor size, degree of di ﬀ erentiation, TNM stage, and CA19-9 levels. Multivariate analyses indicated that advanced TNM stage, excessive ﬁ brinogen, CEA, and CA19-9 levels were independent risk factors for postoperative DFS. And gallbladder neck tumors, poor di ﬀ erentiation, cancer nodules, advanced TNM stage, excessive ﬁ brinogen, and CEA levels were independent adverse factors for postoperative OS. Notably, the preoperative excessive ﬁ brinogen was an independent adverse factor for both DFS ( p = 0 : 044 , HR = 1 : 629 , 95% CI = 1 : 014 − 2 : 618 ) and OS ( p = 0 : 006 , HR = 2 : 328 , 95% CI = 1 : 272 − 4 : 261 ) in patients with gallbladder cancer. The subgroup analyses further indicated that patients with high-level of ﬁ brinogen had both poorer DFS ( p = 0 : 002 ) and OS ( p = 0 : 005 ) than patients with poorly-di ﬀ erentiated gallbladder cancer. And for well-di ﬀ erentiated gallbladder cancer, patients with high ﬁ brinogen levels had poorer OS ( p = 0 : 004 ), but no signi ﬁ cant di ﬀ erence in DFS ( p = 0 : 062 ). Besides, ﬁ brinogen was more signi ﬁ cant in GBC patients with higher BMI ( > 22.62) or older age ( > 60yrs) and was not a ﬀ ected by gender. Conclusion . Elevated preoperative ﬁ brinogen level was independently associated with poor postoperative DFS and OS in patients with gallbladder cancer, especially


Introduction
Gallbladder cancer (GBC) is the most common malignant tumor (about 80% to 95%) in the biliary tract and causes approximately 40,700 deaths worldwide annually in China [1,2]. The prognosis of GBC is extremely poor although it is a rare incident (about 1.2% of all cancers), and the median survival time of all patients is only around 6 months [3][4][5]. Surgery is the main stream intervention for GBC, especially for the radical cholecystectomy with R0 resection has displayed certain benefits in prognosis [6,7]. In principle, simple cholecystectomy could effect a radical cure in patients at T1a assessed preoperatively. For patients at T1b and above, radical cholecystectomy should be performed. Radical cholecystectomy requires additional partial liver resection and regional lymph node dissection, and if necessary, widening the resection to reach R0 margins [8]. However, even in patients with GBC undergoing radical surgery, there is still a high possibility of recurrence (about 25%-65%) after surgery [9,10]. Currently, the absolute survival benefits after preoperative neoadjuvant chemotherapy and postoperative adjuvant chemotherapy with operable GBC are not clear [11,12]. Meanwhile, insufficient evidence has supported the benefit of chemotherapy or radiotherapy in unresectable advanced GBC [13][14][15]. It is difficult to improve the prognosis even with multidisciplinary treatments including reoperation or chemoradiotherapy once GBC recurs. Therefore, a reasonable and effective assessment of recurrence and survival is crucial for postoperative management.
The postoperative survival of patients with GBC is mainly evaluated based on postoperative pathological factors. Studies have shown that postoperative pathological factors such as margin status, TNM stage, histological type, and degree of differentiation of gallbladder cancer were associated with the survival of patients [16]. However, the definitions of TNM stages of GBC varied in four editions of the American Joint Committee on Cancer (AJCC) Staging Manual published from 1997 to 2017, and the definition of regional lymph nodes of GBC in the Japanese TNM staging system was different from AJCC staging system [17,18]. Therefore, pathological factors could not fully reflect recurrence and mortality in patients undergoing radical cholecystectomy.
Hematological markers can assess early disease burden and optimize disease management conveniently, even in preoperation. Representative tumor markers like carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) are commonly used to evaluate the prognosis in malignant tumors [19]. Fibrinogen is a homodimeric glycoprotein composed of 2Aα, 2Bβ, and 2γ peptide chains. In healthy individuals, fibrinogen is mainly synthesized by the liver and circulates at concentrations of 2-5 mg/mL. [20] As the most abundant coagulation factor in plasma, fibrinogen plays an important role in the process of coagulation. During thrombosis, fibrinogen is converted into insoluble fibrin catalyzed by coagulation factors (e.g. thrombin), and the latter forms a fibrin meshwork structure to complete hemostasis [21]. In addition to coagulation and hemostasis, fibrinogen is also a response protein in the acute phase, which is closely related to the progression of inflammation and cancer [22]. It was reported that the elevated fibrinogen before treatment was an adverse prognostic factor in solid tumors, such as breast cancer, lung cancer, and gastrointestinal cancer [23][24][25][26]. It is uncertain about the value of preoperative fibrinogen level to postoperative recurrence and prognostic in patients with GBC. Therefore, developing potential indicators (e.g. hematological markers) will be a good supplement to pathological factors in predicting the prognosis of patients with GBC.
This study retrospectively analyzed 260 GBC patients who underwent radical resection and explored the clinical value of preoperative fibrinogen level in assessing postoperative recurrence and mortality.

Materials and Methods
2.1. General Information. The data of 260 patients with GBC were collected with treatment between January 2015 and December 2020. Among them, 205 cases were from The First Affiliated Hospital, Zhejiang University School of Medicine, and 55 cases were from Zhejiang Cancer Hospital. The inclusion criteria were as follows: underwent treatment of radical cholecystectomy, without any anticancer treatment before surgery, had no other malignancies and/or blood disorders, diagnosed as gallbladder cancer by histopathology, postoperative pathology showed R0 resection margin, and complete clinical data. The ethical approvals were obtained by the Ethics Committees of The First Affiliated Hospital, Zhejiang University School of Medicine (IRB-2014-272) and Zhejiang Cancer Hospital (IRB-2021-276).

Data
Collection. The metadata including gender, age, height, weight, smoking history, drinking history, and body mass index (BMI) were collected by using the hospitals' information system. Meanwhile, preoperative blood indexes and postoperative pathological information were collected. Follow-up data including specific dates of recurrence and death were collected by the hospital system or by telephone.
Serum samples were used for the detection of CEA and CA19-9, and sodium citrate anticoagulated plasma was used for the detection of fibrinogen. All the blood samples were collected from patients in a fasting state within one week before surgery and tested in strict accordance with the instructions of equipment and reagents. And the preoperative hematological indicators contained CEA, CA19-9, and fibrinogen. CEA and CA19-9 were determined by Abbott ALinity i automatic chemiluminescence immunoassay analyzer (USA) and original supporting reagents, and the detection method was chemiluminescence microparticle immunoassay; and Fibrinogen was determined by Sysmex CS-5100 automatic coagulation analyzer (Japan) and original supporting reagents, and the detection method was coagulation method (Clauss method).

Surgical Strategy.
All patients underwent radical surgery for GBC after strict imaging evaluation and the surgery complied with the expert consensus on diagnosis and treatment of gallbladder carcinoma in China [27]. All

Histopathological Examination.
After surgical resection, the tissues/parts of the organ were sent for histopathological examination to find out various histopathological conditions including maximum tumor diameter, tumor location, pathological type, tumor differentiation, the extent of primary tumor invasion, and lymph node metastasis. Pathological examinations were performed by professional pathologists. All the cases underwent TNM staging based on pathological data according to the AJCC Cancer Staging Manual 8 th [28].
2.5. Follow-Up Data. Disease-free survival (DFS) was defined as the time between patients undergoing surgery and cancer recurrence. Overall survival (OS) was defined as the time from patients undergoing surgery to death. Follow-up ended on January 22, 2022, the maximum follow-up period was 80 months and the median follow-up period was 27 (14, 43) months.

Statistical Analysis.
Normally distributed data was expressed as mean ± standard deviation (SD) and nonnormally distributed data was expressed as median (25 th percentile, 75 th percentile) after being tested by the Shapiro-Wilk method. The "survival ROC" package of R Programming Language 4.1.2 was used to calculate the optimal cut-off value of each hematological index by drawing time-dependent receiver operating characteristic (ROC) curves, then the hematological indexes were divided into high-level and low-level groups. Mann-Whitney U test was used to analyze the relationship between postoperative clinical characteristics and fibrinogen. Univariate and multivariate analyses of DFS and OS were performed by using the COX proportional hazards model, which aimed to find independent factors of DFS and OS and calculate their p values, hazard ratios (HRs), and 95% confidence intervals (CI) for HR. And subgroup analyses of fibrinogen in different degrees of differentiation were performed by log-rank test. The above statistics were completed by SPSS 25.0. Statistical results with a two-sided p < 0:05 were considered statistically significant.   (Table 1).

Correlation between Fibrinogen Level and Clinical
Table 1 also showed that preoperative fibrinogen level was associated with age, tumor size, degree of differentiation, TNM stage, and CA19-9 level. Briefly, patients with age>60 (p = 0:017), tumors size >3.5 cm (p = 0:001), poor differentiation (p = 0:013), advanced TNM stage (p < 0:001), and high concentration of CA19-9 (p < 0:001) were higher in fibrinogen levels. The results suggested that a high level of FIB was related to the degree of malignancy and progression of GBC cancer.

Optimal Cut-Off
Values of CEA, CA19-9, and Fibrinogen. As shown in Figure 1, according to the principle of the largest Youden index, the best cut-off values of CEA, CA19-9, and fibrinogen were 2.17 ng/mL, AUC = 0:692; 41.1 U/mL, AUC = 0:694; and 2.97 g/L, AUC = 0:671, respectively. The time-dependent ROC curve, whose expected time was 60 months, decreased the bias caused by different follow-up times. Then CEA, CA19-9, and fibrinogen were grouped based on the cut-off values listed above (Table 1).  (Tables 2 and 3). Notably, multivariate analysis indicated that high-level preoperative fibrinogen was an independent adverse factor for DFS (p = 0:044, HR = 1:629, 95% CI = 1:014 − 2:618) and OS (p = 0:006, HR = 2:328, 95% CI = 1:272 − 4:261) in patients with GBC. In addition to fibrinogen, advanced TNM stage, CEA>2.17 U/mL, and CA19-9 > 41.10 U/mL were independent risk factors for postoperative DFS in patients with GBC (Table 2). While gallbladder neck tumors, poor differentiation, cancer nodules, advanced TNM stage, and CEA>2.17 U/mL were independent adverse factors for postoperative OS in patients with GBC (Table 3).  (Figure 2). All patients were divided into two groups (poor differentiation and well differentiation) and the Kaplan-Meier (logrank test) was used for subgroup analyses. The results showed that patients with high-level fibrinogen were both poorer in DFS (p = 0:002, Figure 3(a)) and OS (p = 0:005, Figure 3(b)) for poorly-differentiated gallbladder cancer. For well-differentiated gallbladder cancer, patients with high fibrinogen levels had poorer OS (p = 0:004, Figure 3(d)) but no significant difference in DFS (p = 0:062, Figure 3(c)). The results suggested that it is best to execute personalized evaluation by combining preoperative fibrinogen with differentiation when assessing DFS in patients with GBC, while OS is not affected by the degree of differentiation.

Discussions
Radical cholecystectomy is still the only way to cure gallbladder cancer although new therapies (e.g. immunotherapy and chemoradiotherapy) have been popularized in clinical practice [29,30]. The gallbladder has a special anatomical structure, which lacks peritoneal coverage on the side facing the liver, so that it makes early GBC also prone to  [31], greatly increasing the recurrence and mortality of GBC patients after radical surgery. Searching for promising biomarkers that affect the recurrence and prognosis of gallbladder cancer from convenient blood parameters is beneficial for the efficient assessment, subsequent therapies, and the postoperative management of GBC patients.
In this study, we demonstrated the preoperative fibrinogen level in patients with gallbladder cancer were correlated to tumor size, differentiation, TNM stage, and CA19-9 level, which indicates the probable promotion of fibrinogen on tumor growth and metastasis. Emerging evidence showed that fibrinogen promotes the metastasis and invasion of cancer cells through multiple pathways. Fibrinogen accelerates  tumor angiogenesis by regulating fibroblast growth factor (FGF-2), thus providing nutrients and gas exchange for cancer cells [32], and ultimately leading to tumor enlargement. In addition, fibrinogen interacts with platelets in plasma to form microthrombi, which act as physical barriers to protect tumor cells from eliminating by natural killer cells, thereby increasing metastatic potential. [33] Studies have shown that fibrinogen enhances cell adhesion through integrin and nonintegrin receptors to bridge vascular endothelium and specific receptors on tumor cells, and promotes tumor cells metastasis to adjacent target organs [34,35]. In poorly differentiated and aggressive tumors, fibrinogen was mediated by tumor development-related proinflammatory cytokines or synthesized and secreted by tumor cells directly [36]. As a consequence, the advanced TNM stage and poor tumor differentiation of GBC patients were causally related to a high level of preoperative fibrinogen. Several publications have confirmed the prognostic value of CEA and CA19-9 in patients with GBC [37,38], while few discoveries were found on the relevance of fibrinogen with gallbladder cancer. This study reveals that high levels of CEA were an independent adverse factor for OS and DFS in GBC patients, and CA19-9 was an independent risk factor for DFS but not OS. Unexpectedly, high-level preoperative fibrinogen was an independent adverse factor both for DFS and OS in patients with GBC, and the hazard ratio of fibrinogen to OS (OS: HR = 2:328 CI: 1.272-4.261) was even higher than CEA (OS: HR = 1:880, CI:1.081-3.271). The results showed that fibrinogen was a better blood index other than tumor markers, which can early assess the recurrence and prognosis of gallbladder cancer. A combination of fibrinogen with pathological factors after surgery can further evaluate the DFS and OS of patients with GBC comprehensively. It was reported CEA ≥3.02 ng/mL was an independent prognostic marker for gallbladder cancer but not CA19-9 [39], which is consistent with our results. The value of CEA in the prediction of the prognosis in gallbladder cancer was confirmed despite the varied cut-off value of CEA. In this study, the time-dependent ROC curves were used to set the cut-off values of fibrinogen, CEA, and CA19-9 at 2.97 g/L, 2.17 ng/mL, and 41.1 U/mL, respectively, which could reduce the bias caused by different follow-up times. It is worthy to verify the cut-off values of the above markers by expanding the sample size. Figure 2 further confirms the value of preoperative fibrinogen to assess postoperative DFS and OS in patients with gallbladder cancer. GBC patients with low preoperative FIB levels (≤2.97 g/L) had no recurrence 3 years after surgery (1-year, 3-year, and 5-year recurrence rates were 23.2%, 39.9%, and 39.9%, respectively). The 3-to 5-year overall survival rate was reduced by only 2% (1-year, 3-year, and 5-year survival rates were 92.6%, 77.6%, and 75.6%, respectively); while the recurrence rate and mortality rate of GBC patients with high preoperative FIB levels (>2.97 g/L) were increasing year by year. Recurrence and death even occur in those patients who have received surgeries more than five years ago. Currently, there is no uniform postoperative monitoring strategy for GBC patients undergoing surgical treatment [40,41]. This study proposes that individualized postoperative management strategies can be formulated according to the preoperative FIB levels of GBC patients. GBC patients with high preoperative FIB levels should be monitored more closely and the follow-up monitoring period should be appropriately extended, while GBC patients with low preoperative FIB level should be appropriately relaxed with the extension of time to save medical resources.
However, subgroup analyses of fibrinogen by differentiation revealed that preoperative fibrinogen was more valuable for recurrence and survival in patients with poorly differentiated gallbladder cancer (Figure 3). There is reason to believe that preoperative fibrinogen can be a good marker for assessing OS of GBC patients regardless of tumor differentiation. When evaluating the DFS of GBC patients, it is best to make a comprehensive judgment based on tumor differentiation. In addition, we observed the suitability of preoperative fibrinogen in different subgroups of age, gender, and BMI. The results found that fibrinogen was more significant in GBC patients with higher BMI (>22.62) or older age (>60 yrs), and was not affected by gender ( Figure S1-S3). If necessary, further comprehensive evaluation can be performed according to the patient's age and BMI. Thence, it is recommended to conduct individualized evaluations according to the clinical characteristics of patients, and formulate more reasonable postoperative management plans. Furthermore, evidence was supporting that inhibiting the pathways of tumor-associated coagulation cascade activation could improve the outcomes of patients [42,43]. As one of the most critical factors in the coagulation pathway, the preoperative fibrinogen level in patients with gallbladder cancer deserves further study.
Studies of gallbladder cancer tend to small sample size, limited by the low incidence. The subjects of this study came from two professional medical institutions, which held relatively large-scale tissue banks and clinical abilities of diagnosis and treatment for cancer patients. However, the study was also with certain limitations. Although the period of the cases in the retrospective analysis is large enough, the selection bias and potential confounding factors in the study cannot be eliminated, and the date of the patient's outcome events cannot be completely accurate. This may be the reason why the area under the ROC curves of CEA, CA19-9, and fibrinogen on the prognosis of gallbladder cancer is not very high. It is necessary to open further prospective studies with a large multicenter sample cohort to solve the issues.

Conclusions
A retrospective analysis of 260 GBC patients who underwent radical cholecystectomy uncovered that patients with high preoperative fibrinogen levels had poor postoperative DFS survival and OS, especially for poor differentiation. Therefore, fibrinogen is probably a potential prognostic biomarker for gallbladder cancer.

Data Availability
Data relevant to the study are available and included in the manuscript.

Ethical Approval
This study was performed according to the Helsinki Declaration. The Ethics Committees of the First Affiliated Hospital, Zhejiang University School of Medicine, and Zhejiang Cancer Hospital approved this study, and the approval number is NO: IRB-2014-272, IRB-2021-276.

Consent
The authors have obtained the informed consents from all participants and their families.

Conflicts of Interest
The authors have no competing interest to declare.

Authors' Contributions
Y.W., J.W., and X.H. were responsible for the study concept and design, data acquisition, and manuscript drafting. S.Z., X.Q., and J.Z. were responsible for data interpretation and statistical analysis. S.Z. and J.Z were responsible for the