The Value of Preoperative Systemic Inammatory Response index (SIRI) Combined with Albumin-Bilirubin Grading (ALBI) in Predicting Early Recurrence Following Curative Resection of Single Hepatocellular Carcinoma

Background: This study was conducted to explore the risk factors for early recurrence of single hepatocellular carcinoma (HCC) following curative resection, and to evaluate the predictive value of systemic inammatory response index (SIRI) combined with the albumin - bilirubin index (ALBI) for early recurrence of HCC following curative resection. Methods: We retrospectively analyzed the clinical data of 233 patients with pathologically conrmed HCC who were admitted to the First Aliated Hospital of Zhengzhou University from January 2015 to January 2018. The optimal cut-off values of alpha-fetoprotein, neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), platelet-to-lymphocyte ratio (PLR), SIRI, and ALBI were determined by receiver operating characteristic curve (ROC), and the predictive eciency of each index was compared by area under curve (AUC). A Cox-proportional hazard regression model was used to analyze factors affecting early recurrence. Kaplan-Meier analysis was used to estimate the 2-year recurrence-free survival (RFS) rate of different factors. Results: The AUC of SIRI for predicting early recurrence was 0.644, which was better than that of NLR, MLR, PLR, and SII. SIRI > 0.96, ALBI >-2.7, tumor diameter > 5 cm, and microvascular invasion (MVI) were risk factors for early recurrence after radical resection of HCC. The AUC of SIRI combined with ALBI was 0.759, which was better than that of single SIRI and ALBI. Combined with the above four risk factors, a new prognostic index was constructed, and the AUC for predicting early recurrence was 0.801. Conclusion: SIRI, ALBI, MVI, and tumor diameter could be considered reliable predictors of early recurrence following curative resection of single HCC. Preoperative SIRI combined with ALBI possesses an important reference value in predicting early recurrence following radical resection of HCC. It is helpful

Tumor-related in ammation and immune responses are essential for the occurrence, development, angiogenesis, and metastasis of malignant tumors [8]. Several studies have shown that preoperative scores based on routine blood in ammatory indicators, such as neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), and platelet-to-lymphocyte ratio (PLR), affect the prognosis of HCC [5,9,10]. In 2016, Qi proposed a new in ammatory index based on peripheral blood lymphocytes, neutrophils, and monocytes, namely the systemic in ammatory response index (SIRI), which has been proven to predict the prognosis of many types of malignant tumors, such as pancreatic cancer, cervical cancer, esophageal gastric cancer, and breast cancer [11][12][13][14]. However, the value of SIRI in predicting early recurrence following radical resection of HCC has not been reported.
Liver function critically affects the treatment and prognosis of patients with HCC [15], it is particularly important to completely evaluate liver function. The Child-Pugh class has been widely applied in the evaluation of liver function and some tumor staging systems; however, the ability to evaluate liver function in patients with HCC with good liver reserve is limited, and is subjectively in uenced by ascites and hepatic encephalopathy [16,17]. Albumin -bilirubin index (ALBI) grade is a newly proposed alternative method for evaluating liver function based on albumin and bilirubin, which can avoid the in uence of subjective factors and can further divide Child-Pugh A patients with HCC into two subgroups with different prognoses [18,19]. Some studies have con rmed that the ALBI grade is better than the Child-Pugh grade in predicting the prognosis of patients with HCC [16, 17,20] and proposed to combine it with conventional prognostic models [21,22]. However, there are few studies on the value of ALBI in predicting early recurrence following radical resection of HCC.
This study aimed to explore the risk factors of early recurrence of HCC following radical resection, clarify the value of SIRI combined with ALBI in predicting early recurrence following radical resection of HCC, and construct a new prognostic index (PI) to predict early recurrence and evaluate its effect. We collected the demographic data of the patients, including sex, age, and history of hypertension and diabetes. The laboratory data within one week before surgery, including hepatitis B surface antigen (HBsAg), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (Tbil), albumin (Alb), alpha-fetoprotein (AFP), peripheral blood leukocyte (WBC), neutrophils, monocytes, lymphocytes, and platelets, were obtained. In terms of tumor characteristics, tumor diameter, liver cirrhosis, microvascular invasion (MVI), and Edmondson-Steiner grade of HCC were considered as possible in uencing factors following resection. Criteria for curative resection (1) no macroscopic tumor thrombi were noted in the hepatic vein, portal vein, bile duct, and inferior vena cava (2) In the absence of adjacent organ involvement, portal lymph nodes, or distal metastases (3) The surgical margin and tumor boundary are > 1 cm, or histologic examination of the cross section of the resected liver is free of residual tumor cells, which is a negative surgical margin [2] Criteria of recurrence New intrahepatic or extrahepatic lesions were detected according to imaging examinations, including ultrasound, enhanced computed tomography (CT), magnetic resonance imaging(MRI), contrast-enhanced ultrasound, and positron emission tomography-CT. Early recurrence was de ned as recurrence within 2 years following radical resection of HCC.

Follow-up
Postoperative follow-up was performed via telephone and outpatient review. The start time of follow-up was the start time of surgery, and the end point was the 2-year recurrence-free survival (RFS), which was de ned as the date from the beginning of surgery to the date of tumor recurrence within 2 years. The last follow-up was performed on May 2021.

Statistical analysis
All the analyses were performed using the IBM SPSS Statistics software (version 26). A receiver operating characteristic (ROC) curve was constructed to determine the optimal cut-off value. Continuous variables are presented as mean ± standard deviation, and categorical variables as frequencies (%). A Coxproportional hazard regression model was used to analyze the factors affecting the early recurrence of HCC. ROC curve analysis was used to evaluate the performance in predicting the early recurrence of HCC.
Survival curves were plotted using the Kaplan-Meier method and compared using the log-rank test. Statistical signi cance was set at p < 0.05.

Clinical features
A total of 233 patients were included in this study according to the inclusion and exclusion criteria. There were 184 men (79%) and 49 women (21%), with an average age of 53.7±10.3 y. Before the operation, 193 patients (82.8%) had complicated hepatitis B, and 202 patients (91.8%) had cirrhosis. All the patients had solitary tumors with an average diameter of 5.0±3.0 cm. All the patients had Child-Pugh grade A preoperatively. A total of 113 patients (48.5%) had early recurrence, 105 patients (92.9%) had intrahepatic recurrence, and 8 patients had extrahepatic recurrence, including 4 cases of lung metastasis, 2 cases of kidney metastasis, 1 case of omental metastasis, and 1 case of lymph node metastasis (Table 1).
The optimal cut-off value and area under the curve (AUC) of AFP, ALBI and in ammatory indicators before surgery The ROC curve analysis demonstrated that AFP=35, ALBI=-2.7, SIRI=0.96, PLR=75.36, NLR=2.47 and MLR=0.3 were the optimal cut-off values. According to the threshold value of each indicator, patients with values less than or equal to the cut-off value were divided into the low-value group, while patients with values larger than the cut-off value were divided into the high-value group. The characteristics of the laboratory examination data after grouping are presented in Table 2. The prognostic value of these in ammatory indicators was further analyzed by comparing the AUC. The results demonstrated that the AUC of SIRI was larger than that of NLR, PLR, and MLR, indicating that the prognostic value of SIRI was better than that of NLR, PLR, and MLR (Table 3).  Value of SIRIALBI combination in predicting early recurrence SIRI and ALBI were combined to divide patients into four groups: low SIRI+ low ALBI group, low SIRI+ high ALBI group, high SIRI+ low ALBI group, and high SIRI+ high ALBI group. ROC curves were constructed based on SIRI, ALBI, and SIRI-ALBI (Figure 1). By comparing their AUC, it was found that SIRI combined with ALBI was superior to SIRI and ALBI in predicting early recurrence of HCC (AUC=0.759, p < 0.001) (Figure 1).
Establishing a new PI for predicting early recurrence following curative resection of HCC  Figure 1). Kaplan-Meier curve analysis showed that the 2-year RFS rate of the low-risk group was higher than that of the high-risk group (p < 0.001) (Figure 2).

Discussion
At present, there is no consensus on the cut-off point for the time period for early recurrence following radical resection of HCC. Some clinical practice guidelines and most studies consider 2 years as the time cut-off point to distinguish early recurrence from late recurrence [3,7,[23][24][25][26][27][28][29]. This study demonstrated that most patients (67.7%) showed recurrence within 2 years following surgery, and the risk of recurrence decreased and tended to lever off after 2 years. Therefore, the time of early recurrence was set within 2 years following surgery in this study. Studies  As one of the most common indicators for the diagnosis and prognosis of HCC, AFP still has a controversial predictive value for early postoperative recurrence [26]. We found that AFP had a poor predictive effect on early recurrence of HCC, with an AUC of only 0.576. Multivariate analysis also indicated that AFP could not be used as an independent factor in uencing early recurrence of HCC following curative resection (p > 0.05). This result is consistent with previous studies [5,6,30], whether AFP could be used as a predictor needs to be further veri ed.
In recent years, an increasing number of studies have demonstrated that in ammation plays an important role in the occurrence, development, and metastasis of malignant tumors [13,25,31]. The potential mechanisms could be that neutrophils are mainly concentrated in the peritumoral matrix of liver cancer tissues, [32] and can release angiogenic factors and in ammatory mediators, such as interleukin-1 β (IL-1β), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and reactive oxygen species (ROS). In addition, neutrophils can inhibit the cytolytic activity of immune cells, which promotes the proliferation and metastasis of cancer cells [5,9,10,31,33]. Neutrophils and tumor-associated macrophages modulate the suppression of anti-tumor immunity by inhibiting the immune function of natural killer cells and T cells, leading to malignant progression [5]. Lymphocytes play an anticancer role in host immunity by inducing apoptosis and inhibiting the migration and invasion of cancer cells [10]. Platelets could protect tumor cells from natural killer cell-mediated lysis, and induce rapid activation of neutrophils, thereby promoting metastasis [28,31].
The majority of HCC cases occur in the context of hepatitis and cirrhosis; therefore, in ammation is particularly signi cant in the process of its occurrence and development. Preoperative scoring systems based on peripheral in ammatory indicators, such as NLR, MLR, and PLR, have also been proven to be associated with the prognosis of HCC [5,9,10]. In 2016, Qi proposed a new in ammatory index based on peripheral blood lymphocytes, neutrophils, and monocytes, namely SIRI [11], and it has been proven to predict the prognosis of various malignant tumors, such as pancreatic cancer, cervical cancer, esophageal cancer of the stomach, and breast cancer [12][13][14]; however, its prognostic value in HCC has not been studied. In this study, patients with HCC complicated with hematological diseases and infectious diseases were excluded to prevent them from in uencing routine blood indicators and interfere with the predictive value of early recurrence. The results showed that SIRI, NLR, and MLR could all be used to predict the early recurrence of single HCC following radical surgery, except for PLR. However, only SIRI was an independent risk factor for early postoperative recurrence. ROC curve analysis showed that among these indices, SIRI was of higher value in predicting early recurrence of single HCC following radical resection.
Child-Pugh grade was proposed by Child, Turcotte, and Paugh and used to evaluate the liver function of patients with liver cirrhosis and portal hypertension [34,35]. Since then, it has been widely applied to evaluate the liver function reserve of patients with liver disease and has been applied in various tumor staging systems, including American Joint Committee on Cancer tumor/node/metastasis (AJCC TNM) staging, BCLC staging, and China liver cancer staging (CNLC) staging. The Child-Pugh score system is composed of ve indicators: bilirubin, albumin, PT, ascites, and hepatic encephalopathy, in which ascites and hepatic encephalopathy are highly subjective, and the degree of ascites is correlated with albumin concentration, decreasing the ability to evaluate liver function reserve [16,17]. In addition, with the constant improvement of antiviral therapy and diagnosis of HCC, the number of patients with early HCC with good liver function has increased [15]. In this study, the preoperative Child-Pugh grades of 233 patients were all grade A; therefore, it cannot be applied to predict early recurrence of HCC.
ALBI, which was put forward in 2015, is a new index evaluating liver function in patients with HCC, especially those with good liver function reserve. [16,18]. This index is only composed of bilirubin and albumin, which are both convenient and objective markers. Many studies have proved that its prediction effect is better than that of the Child-Pugh grade. In this study, 233 patients were divided into two groups by ALBI, and the ABLI > -2.7 group indicated that the prognosis was worse, which further con rmed the previous study.
We speculate that the combination of SIRI and ALBI can improve the prediction of early recurrence following curative resection, because the combination can simultaneously evaluate the in ammation, immune status, and liver function of patients with HCC. In this study, SIRI combined with ALBI was superior to SIRI or ALBI alone in predicting early recurrence in patients with single HCC. Patients in the low SIRI-low ALBI group had a lower 2-year recurrence-free survival rate. Therefore, preoperative SIRI combined with ALBI is an important reference value for predicting early recurrence of HCC following radical resection.
At present, there is no consensus on risk strati cation tools for the early postoperative recurrence of HCC.
Except for the AJCC-TNM tumor staging system in the United States, most tumor staging systems are not from surgically treated patients and do not consider MVI [26,36]. This study showed that the area under the ROC curve for TNM staging was only 0.6, which could be due to the Child-Pugh scoring system used for liver function evaluation in the TNM system, while the Child-Pugh grading of 233 patients with HCC in this study was all grade A before surgery; thus, the difference in liver function between the two groups could not be effectively distinguished. In addition, the TNM-tumor staging system lacks evaluation of tumor-related in ammatory indicators. This study combined preoperative SIRI and ALBI to construct a new PI and con rmed that it is of higher value in predicting early postoperative recurrence.
There are certain limitations to this study. First, this study is a single-center retrospective study, which is prone to selection bias. Second, the cut-off value of each index was selected by the ROC curve. Different cut-off values could affect the nal statistical results and warrants further veri cation. Therefore, further multicenter prospective studies should be conducted. Since the study was conducted anonymously, no consent to publication by the patient was required.

Availability of data and materials
The data that support the fundings of this study are available from the corresponding author (Yang Wu, MD, PhD) on request.

Competing interests
All our authors have no con icts of interest to disclose.  Comparison of systemic in ammatory response index (SIRI), albumin-bilirubin index (ALBI), ALBI-SIRI, and prognostic index (PI) in predicting early recurrence. 2-year recurrence-free survival curves of patients with low-risk group and high-risk group.