Introduction

Hepatocellular carcinoma (HCC) is the fifth most common cancer and the second leading cause of cancer-related death in the world [1]. In China, hepatitis B virus (HBV) infection is a crucial cause of HCC, and more than 80% of HCC patients are caused by chronic HBV infection, resulting in a severe disease burden [2, 3]. The prognosis of HBV-HCC is extremely poor, which has been the focus of research [4]. Surgical resection is an effective treatment for early-stage HCC. However, approximately 50% of patients are found at an advanced stage of the disease and miss the opportunity for surgery [5]. In recent years, antiangiogenic drugs or targeted therapy drugs combined with immunotherapy have become an essential treatment for unresectable or advanced HCC. Based on the positive results of the Phase III IMbrave150 trial, the combination of the programmed death-ligand 1 (PD-L1) inhibitor Atezolizumab and the Bevacizumab have become the first-line standard of care for unresectable HCC [6]. Based on the results of the Phase III REFLECT trial, Lenvima (an oral multi-targeted tyrosine kinase inhibitor) was approved by the Food and Drug Administration in 2015 for first-line treatment of unresectable HCC) patients [7]. Lenvatinib has immunomodulatory activity, and compared with the immunodeficient tumor microenvironment, lenvatinib exhibits improved antitumor activity in the immunocompetent tumor microenvironment [8]. Studies have shown that lenvatinib blocks fibroblast growth factor receptor 4, down-regulates tumor PD-L1 levels and Treg differentiation, and improves anti-PD1 efficacy [9]. Preliminary results of several clinical trials have shown that the first-line treatment strategy of Lenvatinib combined with PD-1 inhibitors can improve the survival of patients with unresectable disease [10, 11].

Due to tumor heterogeneity, there is a wide variation in patients’ response to treatment, with only some benefiting from the same treatment strategy. Therefore, it is essential to identify reliable biomarkers to select the population that may benefit from treatment. Inflammatory markers are convenient, fast, inexpensive, and can be dynamically monitored. Studies have shown that immune-inflammatory markers based on peripheral blood neutrophil, lymphocyte, monocyte, and platelet counts are associated with the prognosis of various solid tumors [12, 13]. The neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and other inflammatory markers have shown high value in predicting the prognosis of HCC patients [14, 15]. However, the relationship between inflammatory markers and the combination treatment of lenvatinib and PD-1 inhibitors in unresectable HBV-HCC patients is yet to be explored.

In this study, we conducted a retrospective study of patients with HBV-HCC who were treated with lenvatinib plus a PD-1 inhibitor as the first-line treatment and analyzed the predictive value of inflammatory markers for the clinical outcome of these patients.

Materials and methods

Case collection

We retrospectively analyzed 117 HBV-HCC patients who received dual therapy of a PD-1 inhibitor plus lenvatinib at Beijing Youan Hospital, Capital Medical University from January 2017 to July 2022. The inclusion criteria were as follows: (1) patients diagnosed with HCC by dynamic computed tomography (CT)/magnetic resonance imaging (MRI) or pathology according to the European Association for the Study of the Liver Clinical Practice Guidelines: Management of Hepatocellular Carcinoma; (2) patients who were positive for hepatitis B surface antigen (HBsAg), negative for hepatitis C virus or HIV antibodies; (3) patients who had not received other immunotherapy before starting PD-1 inhibitor treatment; (4) patients did not have recent infections or autoimmune diseases during the treatment; (5) patients were not diagnosed with malignancy other than HCC. The research received approval from the Hospital Ethics Committee and adhered to the ethical principles of the Declaration of Helsinki.

Clinicopathologic variables and data collection

We collected demographic information of patients (including gender and age), medical history (including history of chronic viral hepatitis B, DM, and history of anti-tumor treatment), and clinical pathological features [including blood chemistry, serum AFP levels, vascular invasion, distant metastasis, Barcelona Clinic Liver Cancer (BCLC) stage] by reviewing electronic medical records. Albumin–bilirubin (ALBI) grades were calculated using the system described by Johnson et al. in 2015 [16]. Child–Pugh scores were calculated according to the method created by Child CG and Pugh RN [17]. The modified ALBI (mALBI) grade criteria were calculated according to the method proposed by Masatoshi Kudo [18]. We also collected routine blood test data on neutrophil count, lymphocyte count, platelet count, and monocyte count from patients before immunotherapy and two to three weeks after the first cycle of immunotherapy. The NLR, PLR, Lymphocyte-to-monocyte ratio (LMR), Platelet-to-albumin ratio (PAR), PNR, and SII were calculated according to the following equations.

NLR = neutrophil count [10\(^9\)/L]/lymphocyte count [10\(^9\)/L]

PLR = platelet count [10\(^9\)/L]/ lymphocyte count [10\(^9\)/L]

LMR = lymphocyte count [10\(^9\)/L]/monocyte count [10\(^9\)/L]

PAR = platelet count [10\(^9\)/L]/ albumin [g/L]

PNR = platelet count [10\(^9\)/L]/neutrophil count [10\(^9\)/L]

SII = neutrophil count [10\(^9\)/L] \(\times\) platelet count [10\(^9\)/L]

The main outcome indicator of the study was PFS, defined as the time from the beginning of immunotherapy to disease progression or last follow-up or death from any cause. The secondary endpoint was OS, defined as the time from the beginning of immunotherapy to death from any cause or last follow-up.

Statistical analysis

Continuous variables that followed a normal distribution were presented as mean ± standard deviation, while those not adhering to a normal distribution were expressed as median and interquartile range (IQR). Using the median value of PFS and OS as the cutoff value, PFS and OS were divided into two subgroups, respectively, and the optimal cutoff values for NLR, PLR, LMR, PNR, PAR, and SII were calculated using receiver operating characteristic (ROC) curves. Prognostic factors affecting PFS and OS were analyzed using Cox proportional hazards models. Initially, variables were subjected to univariate analysis. Subsequently, factors with a P value \(< 0.05\) in the univariate analysis were included in the multivariate analysis. Survival curves were plotted using the Kaplan–Meier method, and intergroup comparisons were made using the log-rank test. A P value less than 0.05 was considered statistically significant in all statistical analyses. Statistical analysis was performed using IBM SPSS Statistics version 25.0 and MedCalc version 22.0 software.

Results

Clinical characteristics

During the study period, a total of 117 patients met the study conditions. The baseline clinicopathological features of the patients are shown in Table 1. Ninety patients (76.9%) were younger than 65 years of age, and most were male (n = 92, 78.6%). Twenty-five (21.4%) patients had DM. Seventy-two patients (61.5%) had baseline AFP levels of less than 400 ng/mL. Most patients were BCLC stage C (75.2%). There were 67 cases (57.3%) with vascular invasion and 50 cases (42.7%) with extrahepatic metastasis. ALBI grade of 1, 2, and 3 were recorded in 15 (12.8%), 84 (71.8%), and 18 (15.4%), respectively. Within ALBI grade 2, there were 22 (18.8%) cases with mALBI grade 2a and 62 (53.0%) cases with grade 2b. The Child–Pugh score was categorized as class A in 81 (69.2%) and class B in 36 (30.8%) patients. The most used antiviral drug was entecavir (59 cases, 50.4%), followed by tenofovir alafenamide (32 cases, 27.4%) and tenofovir disoproxil fumarate (26 cases, 22.2%). The most used PD-1 inhibitor was tislelizumab (46 cases, 39.3%), followed by cedilimumab (43 cases, 36.8%) and camrelizumab (28 cases, 23.9%).

Table 1 Baseline clinicopathological features of the patients
Full size table

The optimal cutoff values for predicting the PFS and OS

ROC curve analysis showed that the optimal cutoff values of NLR, PLR, LMR, PAR, PNR, and SII for predicting PFS were 4.20, 127.52, 2.72, 3.13, 65.43, and 539.47, with respective areas under the curve (AUC) of 0.724, 0.658, 0.608, 0.574, 0.627 and 0.706. The optimal cutoff values of NLR, PLR, LMR, PAR, PNR, and SII for predicting OS were 3.56, 88.12, 3.50, 3.13, 31.43, and 303.66, with respective AUC of 0.808, 0.682, 0.689, 0.600, 0.677 and 0.762 (Table 2).

Table 2 The optimal cutoff values of the preimmunotherapy NLR, PLR, LMR, PAR, PNR, SII for predicting the PFS and OS
Full size table

Factors affecting PFS and OS

Univariate analysis of baseline clinicopathological features indicated that AFP (P = 0.03), DM (P = 0.02), NLR (P \(< 0.01\) ), PLR (P \(< 0.01\)), LMR (P \(< 0.01\)), PNR (P \(< 0.01\)), and SII (P < 0.01) were significantly associated with PFS. Multivariable analysis with these factors revealed that AFP \(\geqslant\) 400 ng/mL (HR 1.69; 95% CI 1.11–2.58; P = 0.01) was an independent risk factor of PFS. The PNR \(\leqslant\) 65.43 (HR 0.50; 95% CI 0.30–0.84; P \(< 0.01\)) and the SII \(\leqslant\) 539.47 (HR 0.54; 95% CI 0.30–0.96; P = 0.03) were independent protective predictors of PFS (Table 3).

Table 3 Univariate and multivariate analyses of the factors affecting PFS
Full size table

Univariate analysis of baseline clinicopathological features indicated that AFP (P = 0.04), DM (P = 0.04), Extrahepatic metastasis (P = 0.03), NLR (P \(< 0.01\)), PLR (P \(< 0.01\)), LMR (P = 0.03), PAR (P = 0.02), PNR (P \(< 0.01\)), and SII (P \(< 0.01\)) were significantly associated with OS. Multivariable analysis with these factors revealed that AFP\(\geqslant\) 400 ng/mL (HR 1.97; 95% CI 1.22–3.17; P \(< 0.01\)) and DM = yes (HR 2.43 95% CI 1.38–4.27; P = 0.02) were independent risk factors of OS. The SII \(\leqslant\) 303.66 (HR 0.40; 95% CI 0.19–0.85; P = 0.02) was an independent protective predictor of OS (Table 4).

Table 4 Univariate and multivariate analyses of the factors affecting OS
Full size table

Kaplan–Meier survival analyses showed the relationship between SII and survival results. The patients with an SII of \(> 539.47\)attained poorer PFS (HR = 2.56; 95% CI 1.66–3.94; P \(< \,0.001\)) (Fig. 1) compared with the patients with an SII of \(\leqslant\) 539.47. The patients with an SII of \(> 303.66\) showed a significantly worse OS than those with an SII of \(\leqslant\) 303.66 (HR 2.78; 95% CI 1.80–4.28; P \(< \,0.001)\) (Fig. 2).

Fig. 1
figure 1

Kaplan–Meier analysis of the PFS according to the preimmunotherapy SII, with log-rank test (cutoff: 539.47)

Full size image
Fig. 2
figure 2

Kaplan–Meier analysis of the OS according to the preimmunotherapy SII, with log-rank test (cutoff: 303.66)

Full size image

The impact of changes in SII before and after immunotherapy on PFS and OS

Kaplan–Meier survival analyses showed that after the first cycle of immunotherapy, the patients with a increased SII attained poorer PFS (HR 1.61; 95% CI 1.10–2.37; P = 0.015) (Fig. 3) and OS (HR 1.76; 95% CI 1.15–2.70; P = 0.009) (Fig. 4) compared with the patients with a decreased SII.

Fig. 3
figure 3

Kaplan–Meier plots of PFS by changes in SII

Full size image
Fig. 4
figure 4

Kaplan–Meier plots of OS by changes in SII

Full size image

Association between SII and therapeutic response

According to modified Response Evaluation Criteria in Solid Tumors (mRECIST) guidelines, one patient (0.8%) achieved complete response (CR), 43 (36.8%) achieved partial response (PR), 47 (40.2%) achieved stable disease (SD) and 26 (22.2%) achieved progressive disease (PD). The corresponding objective response rate (ORR) was 37.6%. The ORR of the patients whose SII had decreased (47.5%) was higher than patients whose SII had increased (32.5%), but the difference was not statistically significant (P = 0.11). We compared the therapeutic responses of the two groups using the Mann–Whitney test and found that there was a significant difference between the SII-increased patients and the SII-decreased patients (Z = – 2.055, P = 0.04) (Table 5).

Discussion

Globally, liver cancer is one of the most common malignant tumors, with HBV-HCC being the most common type of liver cancer [19]. In recent years, with the rapid development of molecular targeted therapy and immunotherapy, patients’ survival has greatly improved [20]. The strategy of combining targeted drugs with immunotherapy has become one of the most important first-line treatments for advanced HCC. Research on Immune checkpoint inhibitors (ICIs) combined with antiangiogenic drugs has shown promising prospects. Antiangiogenic drugs represented by lenvatinib combined with anti-PD-1/PD-L1 antibodies have shown encouraging ORR in advanced HCC patients. Lenvatinib combined with nivolumab and pembrolizumab achieved ORRs of 54.2% and 36.0%, respectively [10, 21]. The latest Phase III LEAP-002 study further suggests that lenvatinib combined with pembrolizumab provides a survival benefit compared to lenvatinib alone (median OS of 21.2 months vs. 19.0 months) [11]. However, there are significant individual differences between immunotherapy and targeted therapy. Studies have shown that the response to immunotherapy differs between hepatitis B virus-related and non-viral HCC [22]. Therefore, finding reliable predictive biomarkers has always been a major challenge in clinical practice.

HBV-HCC arises in the context of chronic inflammation and within a fibrotic liver and is an inflammation-driven carcinoma [23]. Increasing evidence suggests that inflammatory responses are associated with the efficacy of anti-PD-1 therapy in cancers, and systemic inflammatory markers can be used to determine prognosis [24, 25]. However, the inflammatory biomarkers that best predict prognosis in anti-PD-1 therapy, especially in HBV-HCC patients, remain unclear. Many studies in hematological markers have shown that NLR can reflect inflammation and host immune response, and high NLR level before treatment is a poor prognostic factor for patients with a variety of malignant tumors receiving immunotherapy [26,27,28] (27–29). In HCC, this idea was also confirmed [29]. Although NLR can help to assess the prognosis of immunotherapy, it only integrates the two cell types. SII combines the three key indicators of neutrophil, lymphocyte, and platelet to reflect the complex immune response and inflammatory state. Can it better predict the prognosis of patients? Our study found that the SII outperformed other inflammatory markers regarding prediction. Our findings are consistent with previous studies. Irsan Hasan et al. [30] found that when evaluated for their role as predictors of 1-year survival in untreated patients with advanced HCC, the NLR had a discriminatory ability based on the area under receiving operator curve of 0.667 (95% CI 0.536–0.798; P = 0.044), with the SII of 0.766 (95% CI 0.643–0.889; P = 0.001). SII had superiority in discriminatory ability (p = 0.0415). Yang et al. [31] also reported that an increase in SII with a cutoff value of 300 was negatively associated with OS in HBV-HCC, and in our study, this cutoff value was 303.66. Another study in patients with advanced HCC treated with PD-1 inhibitors found that NLR at six weeks was a predictor of PFS and OS, whereas SII at six weeks was only a predictor of PFS [32]. The study included patients receiving both anti-PD-1 combination therapy and ICI monotherapy, which may lead to differences in the findings. In conclusion, SII is an effective biomarker and results in better survival with low SII compared with high SII in HBV-HCC patients receiving immunotherapy. Although SII can reflect a patient’s response to treatment, few studies have focused on the dynamic changes of SII in HBV-HCC patients. Our study shows for the first time that SII decrease after treatment is significantly associated with prolonged PFS and OS in HBV-HCC patients receiving first-line lenvatinib plus a PD-1 inhibitor.

DM is widely recognized as a risk factor for cancer. Studies have shown that DM increases the risk of liver cancer [33], and HCC patients with DM have a higher risk of tumor progression [34]. The mechanism by which DM leads to the progression of HCC is multifaceted, involving inflammation, insulin resistance, abnormal lipid metabolism, oxidative stress, and other aspects [35]. In addition, in our study, AFP \(\geqslant\)400 was also established as an independent risk factor for PFS and OS, consistent with the results of previous studies [36, 37]. The combination of AFP and SII can better predict the prognosis of such patients.

This study has several limitations. First, it is a small-scale, single-center clinical study with a small sample size, which may lead to selection bias and insufficient representativeness of the research results. Future research should focus on validating SII in larger cohorts of patients, exploring their interactions with other prognostic factors, and elucidating the precise mechanisms underlying their prognostic significance.

Table 5 Therapeutic efficacy of response
Full size table

Conclusion

In routine clinical practice, SII is a simple and practical blood marker that can help predict the prognosis of HBV-HCC patients treated with first-line lenvatinib plus PD-1 inhibitors.