In the past few decades, hepatocellular carcinoma (HCC) has risen to become the sixth most aggressive malignancy and the fourth-leading cause of cancer-related deaths worldwide (1). Among the various therapeutic approaches, hepatic resection (HR) is presently considered one of the most curative modalities for HCC patients (2). With the continuous improvement of perioperative management and operation techniques over the past few years, although the overall survival time of the patient has been prolonged, the recurrence of HCC after hepatectomy remains high (3). As reported by the literature, the cumulative 5-year recurrence rate for HCC patients after hepatic resection was approximately 80% in most centers (4, 5).

The potential mechanism of postoperative recurrence could be an intrahepatic relapse of the primary tumor (considered as early recurrence) or the emergence of de novo multicentric tumor (stated as late recurrence). Therefore, according to the time to postoperative recurrence, we artificially divided the recurrence of liver cancer into early recurrence (recurrence within 24 months) and late recurrence (recurrence over 24 months) (6–9). Numerous recent studies have shown that early recurrence is related to biological characteristics of tumors such as tumor size, tumor numbers, tumor differentiation, microscopic vascular invasion (MVI), and satellitosis (10–12). There is a lack of data to predict late recurrence. Scholars tend to accept the view that late recurrence is commonly associated with the intrinsic liver condition, like liver cirrhosis (13, 14).

Liver cirrhosis is closely related to portal hypertension (PH) and PH could influence the natural history of the underlying liver condition in return (15). We couldn’t accurately assess PH clinically without some invasive methods. The spleen is a sensitive organ to PH alteration and responds promptly in the form of splenomegaly. Moreover, splenomegaly can contribute to the progression of PH by increasing splenic blood flow (16). However, the relationship between splenomegaly and liver cirrhosis or PH is controversial. This is partly because the spleen is irregular in shape and the spleen size is measured by longitudinal spleen axis on ultrasound or CT images, which makes it not as accurate as three-dimensional measurement.

Recently, liver computed tomography (CT) volumetry has been used to help assess liver volume in clinical practice (17, 18). Especially, when considering the future liver remnant (FLR) volume, we can use three-dimensional CT reconstruction to accurately represent the remaining liver volume (19). Similarly, with the same use of three-dimensional CT volumetry software, the splenic volume could be feasible to obtain. Some studies have shown that spleen volume measured by CT is a dependable indicator for liver fibrosis, cirrhosis, and portal hypertension (20). As far as we know, the role of splenic volume as a predictor of HCC late recurrence has never been explored. Therefore, this study aimed to assess whether the SV measured by 3D reconstruction could predict HCC late recurrence after hepatectomy.

Although surgical resection is beneficial to HCC patients, the high recurrence rate of HCC patients after hepatectomy remains the most serious challenge for both clinical surgeons and patients (9). Therefore, to better improve the management of these patients, identifying predictors of recurrence after hepatectomy is indispensable.

The determination of cut-off time between the early and the late recurrence has always been a focus of debate because of their distinct physiopathologic mechanism and strategies of surveillance, prevention, and management of HCC recurrence. Although a few studies (22, 23) suggested that less than 1 year from resection to recurrence has a worse prognosis, the cut-off of 2 years after hepatectomy has been widely accepted by researchers (6–9). Our study that there are completely different independent factors between early and late recurrence may indirectly support this point of view.

The risk factors for early HCC recurrence are mainly tumor-related, which is consistent with our study. In our study, we found that AFP (>400ng/ml), MVI, satellitosis, and BCLC staging were independent risk factors of HCC early recurrence through the multivariable Cox regression analysis. AFP serves as a surrogate marker for tumor differentiation as well as vascular invasion and high-level AFP was reported to be linked to HCC recurrence (24, 25). Microvascular invasion (MVI) is defined as the presence of tumor cell clusters in the branch of the portal or hepatic vein under microscopy (26). Obviously, the presence of MVI signifies the invasion of tumor cells into the vessel. Roayaie et al. suggested that MVI is an extremely important factor that can accurately predict the risk of recurrence (27). As for satellitosis, it refers to small metastases around the main tumor. In general, satellitosis originated from MVI, indicating that tumor metastasis has entered a late pathological stage (28). Several articles (29, 30) revealed that HCC patients with satellitosis had a worse overall survival than those without satellitosis. Our study further confirmed that satellitosis is a risk factor for early recurrence. Among many liver cancer staging systems, The Barcelona Clinic Liver Cancer (BCLC) system is currently the most widely applied system around the world. The unique advantage of BCLC staging is that the patient’s performance status, tumor burden, and liver function are considered comprehensively (31, 32). Advanced-stage HCC (BCLC-C HCC) patients have tumors that have spread beyond the liver, as well as vascular invasion. Similarly, BCLC staging has been widely validated and associated with the prognosis of HCC patients by several multicenter studies (33, 34).

Subsequently, patients without early recurrence were accepted into our further analysis for late recurrence. After multivariate analysis, splenic volume was the only predictor of HCC late recurrence. The curves for late HCC recurrence-free survival showed a statistically significant difference: patients with SV ≥165 ml had a higher recurrence rate. And for patients with SV <165 ml, fewer than half of them relapsed during the study period. Through comparison between the low SV and the high SV groups, we found that most tumor-related variables (tumor size, AFP, MVI, BCLC staging, and tumor differentiation) didn’t add a contribution to late recurrence. So according to the cut-off of 165 ml, we could preoperatively classify patients as high or low late recurrence of the crowd based on actual splenic volume (after excluding early recurrence-related factors). To obtain the late recurrence rate more accurately, we developed the nomogram based on SV. Depending on the specific value of SV, we could get the likelihood of the 3-year RFS, 4-year RFS, and 5-year RFS, which greatly facilitates individualized clinical surveillance and prevention of HCC recurrence.

There is no doubt that the risk factors for early HCC recurrence are biological characteristics of the tumor (10–12). Whereas, there are poor data for predicting late recurrence (35). It is widely acknowledged that the severity of chronic liver conditions (like liver cirrhosis) plays a significant role to some degree (13, 14). It is proven that the severity of portal hypertension accelerates the natural process of chronic liver disease, including HCC occurrence (36). Admittedly, the recurrence of HCC after 2 years of surgery was considered as a result of the development of de novo tumors or new malignant clones (9), which additionally confirms that late recurrence is correlated to the underlying liver condition. Therefore, we could assess the degree of liver cirrhosis by calculating the level of portal hypertension. In clinical, the Hepatic Venous Pressure Gradient (HVPG) is viewed as the gold standard for diagnosing PH. Due to its invasion, many scholars were devoted to searching for a non-invasive and simple method to evaluate HVPG. Both spleen size and platelet count were reported to serve as surrogate markers in assessing HVPG (37). In our univariate analysis for late recurrence, splenic volume (P=0.001) and platelet count (P=0.038) are risk factors, which supported previous theoretical research. It is well known that thrombocytopenia is caused by hypersplenism secondary to splenomegaly and is also one of the complications of portal hypertension. In our further multivariate analysis, however, the splenic volume (P=0.001) may play a more dominant role than platelet count (P=0.760).

HCC patients are often accompanied by congestive splenomegaly due to the background of cirrhosis. It is often seen in the increase of splenic vein pressure caused by the increase of portal vein pressure, which makes it difficult for the blood of the spleen to return, and then causes the blood stasis of the spleen, resulting in splenomegaly (38). In other words, Splenomegaly is regarded as an important symbol of cirrhosis and portal hypertension. Takeishi et al. showed that splenic volume could help predict HCC prognosis after hepatectomy, and combined splenectomy with hepatectomy can reduce tumor recurrence in patients with splenomegaly (39). Some studies (40, 41) have also found that HCC patients with splenomegaly tend to have a poor liver function. And for those patients, the incidence of tumor-related complications was significantly increased after surgeries, leading to a poor prognosis. It has also been discovered that the spleen plays an essential role in immunity. In Hashimoto’s article (42), patients with splenomegaly have a higher number of CD4+ regulatory T cells, PD-L1, and PD-L2 expression cells than those without splenomegaly, implying that patients with splenomegaly have poor tumor immunity. The inhibitory influence of CD4+ Treg cells on immune function is mainly reflected in the inhibition of activation and proliferation of helper T cells and cytotoxic T cells, and it has been found that CD4+ Treg cells are involved in the formation of tumor immune tolerance and the inhibition of anti-tumor immune response (43). Similarly, cells expressing both PD-L1 and PD-L2 are crucial in the mechanism of tumor immune escape of the tumor microenvironment. In conclusion, decreased tumor immunity caused by splenomegaly may be one of the reasons for liver cancer recurrence.

For patients with large SV, liver transplantation has incomparable advantages compared with other treatment methods. It is the only method that can completely cure cirrhosis, HCC, and portal hypertension. However, because liver transplantation has high requirements for doctors, hospital equipment and conditions, as well as patient economy, not all patients have conditions to accept liver transplantation first. For those patients who cannot accept liver transplantation, the treatment of hepatocellular carcinoma should be given priority (re-resection). Therefore, for those patients with large SV, if they cannot accept primary liver transplantation (PLT), we’ll consider hepatectomy first. Salvage liver transplantation (SLT) could be performed after several times of HCC recurrence or liver failure. It was shown that overall survival and disease-free survival were the same after PLT or SLT (44).

Our study makes use of splenic volume to predict HCC late recurrence. Patients only need a preoperative CT scan, and SV is automatically obtained by using three-dimensional CT volumetry software. Another advantage of using SV as a predictor is that in comparison to liver volume, splenic volume in adults is not affected by age, BMI, or other factors (45).

However, the current study has several limitations. Firstly, the possibility of selection bias certainly exists because of its retrospective nature. All data come from a single center and the late recurrence group had a small number of patients, which might make our results less convincing. Secondly, for developing the nomogram, the perfect design is to set up a training cohort and use other data to set up a validation cohort. This is really the shortcoming of our experiment. In fact, building a nomogram is not the main point of our article. Our team did not spend a lot of words describing the nomogram. Also, three-dimensional reconstruction of the spleen is not common in other hospitals. We have difficulty getting patient data from other hospitals as a validation cohort. Thirdly, we ignored the effect of anti-tumor therapy, such as antiviral drugs and neo/adjuvant systemic therapy, on the recurrence of liver cancer after surgery because we couldn’t control the duration time and dose of therapy used by patients. A future multicenter prospective study is necessary.

In conclusion, our study verified that tumor-related factors are predictors of HCC early recurrence and splenic volume was related to HCC late recurrence because of its association with cirrhosis and PH. With the increase of SV, the probability of late recurrence also increases. This finding could assist us in guiding the clinical surveillance and prevention of HCC recurrence.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The study was approved by the ethics committee of Xiangya Hospital of Central South University.

TF and LZ contributed to conception and design of the study. TF and LM organized the database. TF and GL performed the statistical analysis. TF wrote the first draft of the manuscript. TF wrote sections of the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the grants National Nature Science Foundation of China (NO. 81771932).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.