Association between periodontitis and hepatocellular carcinoma across different tumor stages

Abstract

Objectives

Hepatocellular carcinoma (HCC) is a malignant liver cancer. While periodontitis associates with many chronic liver diseases, oral health data of HCC patients remain limited. This study assessed dental/periodontal health differences across different stages and confirmed the association between periodontitis and HCC.

Materials and methods

113 HCC patients and 30 healthy volunteers were enrolled. HCC patients were stratified by China Liver Cancer Staging (CNLC): Ⅰa, Ⅰb, Ⅱa, Ⅱb, Ⅲa, and Ⅲb. Biochemical tests and oral examinations data were collected at HCC patients’ admission and analyzed to their oral characteristics and explore correlations between periodontitis and various liver-related indicators.

Results

Compared with healthy individuals, HCC patients exhibited significantly increased periodontal probing depth (PD), gingival recession, clinical attachment loss (CAL), bleeding on probing (BOP), and higher tongue disease prevalence. Additionally, stage Ⅲb patients had the most severe periodontitis. Logistic regression identified PD, tooth loss, and BOP as HCC-associated factors. Correlation analysis showed that PD was significantly correlated with aspartate transaminase, alanine transaminase, fibrinogen, D-dimer, alpha-fetoprotein, and protein induced by vitamin K absence or antagonist-II. BOP was significantly correlated with aspartate transaminase, alkaline phosphatase, γ-glutamyl transpeptidase, activated partial thromboplastin time, and D-dimer (all P < 0.05).

Conclusions

Periodontitis is associated with HCC. PD and BOP correlate with key HCC markers. These findings emphasize the relationship between periodontitis and HCC and highlights the need for interdisciplinary collaboration between dentistry and hepatology to optimize oral care and overall management of HCC patients.

Trial registration

This study has been registered at Chinese Clinical Trial Registry (ChiCTR2400082256, March 25, 2024, https://www.chictr.org.cn/).

Introduction

Hepatocellular carcinoma (HCC) is a malignant liver cancer. Liver cancer ranks 7th in terms of incidence among all malignant tumors, and it is also the second leading cause of cancer-related deaths. This poses a major threat to socioeconomic development and human lives [1, 2]. Specifically in China, there are over 460,000 new cases of liver cancer each year, accounting for more than half of the global new cases. Among all primary liver cancer cases, HCC accounts for 75%-85% [3].

Periodontitis is a highly prevalent chronic inflammatory disease caused by periodontal pathogenic bacteria infection. This local inflammation not only destroys periodontal supporting tissues but also induces the translocation of oral pathobionts (mainly periodontal pathogens) and their metabolites into the systemic circulation, thereby triggering persistent low-grade systemic inflammation [4]. In recent years, a growing body of studies has established a clear link between oral and periodontal diseases and liver diseases, emphasizing that oral health status is an important influencing factor of liver health [5]. Firstly, periodontitis-induced systemic inflammation can directly act on the liver, inducing liver injury and participating in the progression of chronic liver diseases (CLDs). Secondly, the translocation of pathogenic bacteria and endotoxins from the oral cavity to the systemic circulation leads to endotoxemia, which further aggravates liver inflammation and damage [6]. Moreover, periodontitis can induce gut dysbiosis. This disrupted gut microbiota enhances intestinal barrier permeability, promoting intestinal translocation of harmful substances, which in turn affects the pathophysiology of CLDs through the gut-liver axis [7].

The association between periodontal disease and some CLDs (such as hepatitis and cirrhosis) is relatively clear. But data on oral health of HCC patients has not been extensively reported. Moreover, the treatment options for HCC depend on the disease stages [8]. Therefore, the health status of HCC patients at different stages is worthy of great attention. Thus, this study focused on exploring the oral and periodontal health status of HCC patients at different stages based on China Liver Cancer Staging (CNLC). Notably, epidemiological studies found a positive correlation between periodontitis and overall cancer risk [9]. And a cohort study had further suggested an association between severe periodontitis and increased mortality risk from oral and digestive system cancers, including HCC [10]. Accordingly, we hypothesized that the severity of periodontitis was associated with the progression of HCC staging. Given that HCC is one of the leading causes of global cancer-related mortality and periodontitis is a highly prevalent modifiable disease, this study aimed primarily to investigate the association between periodontitis and HCC across different CNLC stages, which can help formulate targeted preventive strategies and improve the prognosis of HCC patients. Moreover, this study was anticipated to provide enhanced support for the interdisciplinary management of HCC patients.

Materials and methods

Study design

This study was conducted at the First Affiliated Hospital of Xi’an Jiaotong University in Xi’an, Shaanxi Province, China, in adherence to the ethical principles of the Declaration of Helsinki. The study protocol was approved by the hospital’s Institutional Review Board (No. XJTU1AF2024LSYY-057) prior to initiating data collection and analysis. Written informed consent was obtained from all participants before their enrollment in the study.

Participants

HCC patients were eligible for inclusion if they met the following criteria: (1) aged ≥ 18 years; (2) histologically confirmed HCC via liver biopsy (where feasible), or a definitive diagnosis established by a combination of clinical manifestations and radiological findings; (3) clear consciousness; and (4) willingness and ability to cooperate with all study-related assessments. The exclusion criteria were as follows: (1) metastatic hepatic carcinoma; (2) a history of primary or secondary malignancies other than HCC; (3) receipt of any dental treatment within 3 months prior to study enrollment; (4) pregnancy or lactation; and (5) failure or refusal of the patient or their legal surrogate to provide written informed consent.

The healthy control participants were enrolled based on the following inclusion and exclusion criteria. Inclusion criteria were defined as: (1) aged ≥ 18 years; (2) free of any subjective symptoms of liver diseases; (3) clear consciousness; and (4) willingness and ability to cooperate with all study-related assessments. Exclusion criteria included the following: (1) a history of CLDs; (2) a personal history of any malignant tumor, including HCC and other primary or secondary malignancies; (3) presence of severe systemic diseases (e.g., autoimmune diseases, renal insufficiency); (4) receipt of any dental treatments within 3 months prior to study enrollment; (5) long-term administration of medications that might affect liver function or periodontal status (e.g., immunosuppressants, glucocorticoids, chemotherapeutic agents); (6) pregnancy or lactation; and (7) presence of mental disorders or inability to complete the required study procedures.

All patients were grouped according to their CNLC stages at admission (CNLC staging criteria were shown in Fig. 1). According to the 2024 Edition of Diagnosis and Treatment Standards for Primary Liver Cancer, liver cancer can be divided into four stages: Stage I: There is a single tumor, with no vascular invasion or extrahepatic metastasis, and the Child-Pugh grade is A or B. Stage II: There are 2–3 tumors, each with a maximum diameter exceeding 3 cm, and there is no vascular invasion or extrahepatic metastasis either. Stage III: The tumor has vascular invasion, even if there is no extrahepatic metastasis. Stage IV: The Child-Pugh grade is C [11].

Fig. 1.

The China Liver Cancer Staging (CNLC) for hepatocellular carcinoma patients (HCC). All patients with HCC will be staged based on physical activity status score, Child-Pugh, have extrahepatic metastasis or not, have vascular cancer thrombus or not, and the number and size of tumors

Assessments

Demographic characteristics

The demographic characteristics of all participants were documented, including gender, age, body mass index (BMI), smoking status, and alcohol consumption history.

Blood collection and laboratory testing

Venous blood samples were collected from all patients with HCC after overnight fasting for at least 8 h. A total of 5 mL of peripheral venous blood was drawn using sterile vacuum blood collection tubes, including 3 mL in coagulation tubes (containing sodium citrate anticoagulant) and 2 mL in serum separation tubes. The collected blood samples were transported to the clinical laboratory within 2 h. All laboratory indicators were detected using automated biochemical analyzers (Beckman Coulter AU5431, America) and coagulation analyzers (Stago STR, France) following the instrument operating specifications and reagent kit requirements. The brief detection methods are as follows: (1) Liver function-related indicators: Aspartate transaminase (AST), alanine transaminase (ALT), alkaline phosphatase (ALP), γ-glutamyl transpeptidase (GGT), cholinesterase (CHE), total bilirubin (TBIL), direct bilirubin (DBIL), and indirect bilirubin (IDBIL) were detected by the enzymatic method with corresponding reagents, and their concentrations were calculated via standard curves. (2) Coagulation function-related indicators: Prothrombin time (PT) and prothrombin time activity (PTA) were detected by the one-stage method; activated partial thromboplastin time (APTT) by the kaolin-cephalin activation method; thrombin time (TT) by the thrombin substrate method; fibrinogen (FIB) by the clotting turbidimetry method. All were tested with anticoagulated blood, and PTA was derived from PT values. (3) Thrombus-related and tumor marker indicators: D-dimer (D-D) was detected by immunoturbidimetry. Alpha-fetoprotein (AFP) and protein induced by vitamin K absence or antagonist-II (PIVKA-II) were detected by chemiluminescence immunoassay, and their concentrations were quantified via standard curves.

Oral clinical examination

Oral health status of each subject was evaluated via a full-mouth clinical examination, which involved inspection of the teeth, periodontal tissues, and tongue.

Decayed, missing, and filled teeth (DMFT) and periodontal alveolar bone defects of all subjects were assessed using Cone beam computed tomography (CBCT). A CBCT machine (KaVo 3D eXam; KaVo Dental, Bismarckring) was used to acquire 3D images with the following parameters: 120 kV, 37.1 mA, 23 × 17 cm field of view, 0.3 mm voxel size, 17.8 s exposure duration, and 0.3 mm slice thickness. Prior to imaging, interdisciplinary consultations involving oral radiologists and hepatologists were conducted to confirm that the low radiation dose of CBCT is far below the threshold for inducing adverse biological effects in humans and thus ensures the safety of this imaging modality for both healthy participants and patients with HCC. Periodontal alveolar bone defects were classified according to the American Academy of Periodontology (AAP) wall classification system: one-wall defect, two-wall defect, three-wall defect, and four-wall defect. Following the CBCT scan, the data were collected and converted to the Digital Imaging and Communications in Medicine file format.

A periodontal examination using a periodontal probe was performed on all subjects to find the differences in periodontal probing depth (PD), gingival recession, clinical attachment loss (CAL), and bleeding on probing (BOP) between healthy volunteers and HCC patients. The periodontal probe was guided along the long axis of teeth on various surfaces, measuring PD at 6 sites per tooth (proximal, central, and distal regions of buccal and lingual surfaces) via an insertion motion along the periodontal pocket base. The probe was moved along the bottom of the periodontal pocket in a plug-in manner during the probing process.

PD was recorded by measuring the distance from the gingival margin to the periodontal pocket base using the probe scale. Gingival recession was measured by the scale of the periodontal probe, and the maximum value of gingival recession was recorded for each tooth. CAL was recorded as the distance from the gingival sulcus/periodontal pocket base to the cementoenamel junction (CEJ). After measuring the depth of the periodontal pocket, the CEJ position was identified as the probe tip slid along the root surface, and the distance from the CEJ to the gingival margin was measured. CAL was calculated as follows: PD minus the CEJ-to-gingival margin distance (if the result ≥ 0 or CEJ was undetectable, CAL was considered 0); sum of PD and the CEJ-to-gingival margin distance (if gingival recession placed the gingival margin apical to the CEJ). The results of PD, gingival recession, and CAL were calculated as the mean of all teeth in the mouth. CBCT provided three-dimensional imaging data to observe alveolar bone defects. Besides, the position of CEJ was verified through CBCT to avoid significant errors or recording errors and improve accuracy.

BOP was recorded by gently inserting the probe into the gingival sulcus/pocket base, removing it, and observing for bleeding after 10–15 s. The results of BOP were calculated as the percentage of bleeding sites to all sites in the mouth.

The status of the oral mucosa was assessed through visual examination and palpation, with particular focus on checking for any pathological changes in the tongue. Recorded findings included fissured tongue, atrophic glossitis, swollen tongue papillae, hairy tongue, and ulcers.

Statistical analysis

Statistical analysis was conducted using GraphPad Prism 9 software (GraphPad Software Inc.) and SPSS version 31.0 software (IBM Corp.). The D’Agostino-Pearson test was used to conduct a normality test on all data. For normally distributed data with homogeneous variance, one-way analysis of variance (ANOVA) was used to identify overall significant differences, followed by Tukey’s multiple comparisons test to pinpoint group-specific differences. For non-normal or heteroscedastic data, the Kruskal-Wallis H test was used, and Dunnett’s multiple comparisons test was used to compare the differences between groups. In all statistical comparisons, a two-tailed P < 0.05 was considered statistically significant.

Results

Characteristics of all participants

This study included 113 HCC patients. At the same time, 30 healthy volunteers were included as a control group. The mean patient age was 60.3 (standard deviation [SD]: ± 10.9) years. And 83% of patients were male. HCC patients were divided into six groups based on the CNLC stage of HCC: stage Ⅰa (n = 24), stage Ⅰb (n = 15), stage Ⅱa (n = 19), stage Ⅱb (n = 14), stage Ⅲa (n = 26), and stage Ⅲb (n = 15). There was no statistically significant difference in gender ratio and age among all groups (Table 1).

Table 1.

Characteristics of teeth and periodontal indicators in control group and HCC patients

	Control	HCC	CNLC stage of HCC
	n = 30	n = 113	Ⅰa	Ⅰb	Ⅱa	Ⅱb	Ⅲa	Ⅲb
			n = 24	n = 15	n = 19	n = 14	n = 26	n = 15
Age, years	62.6 (9.6)	60.3 (10.9)	56.3 (9.9)	65.2 (12.7)	60.2 (12.6)	65.4 (11.2)	59.2 (9.0)	58.8 (9.0)
Gender
Male, n (%)	25 (83%)	94 (83%)	20 (83%)	12 (80%)	15 (79%)	12 (86%)	22 (85%)	13 (87%)
Female, n (%)	5 (17%)	19 (17%)	4 (17%)	3 (20%)	4 (21%)	2 (14%)	4 (15%)	2 (13%)
Teeth (n)	27.1 (1.2)	24.0 (5.5)	27.0 (1.9)	26.2 (2.5)	22.2 (6.0) **##	22.2 (7.5) *#	23.5 (5.5) **##	21.9 (6.9) **##
Lost teeth (n)	0.9 (1.2)	4.0 (5.5)	1.0 (1.9)	1.8 (2.5)	5.8 (6.0) **##	5.8 (7.5) *#	4.5 (5.5) **##	6.1 (6.9) **##
Decayed and filled teeth (n)	1.8 (1.6)	4.1 (4.3)	1.8 (2.3)	5.2 (6.0)	4.9 (4.8)	5.3 (4.6)	5.0 (4.2) #	2.7 (2.7)
Type of bone defect (n)
One-wall defect	0	0	0	0	0	0	0	0
Two-wall defect	0	0.06 (0.31)	0	0	0.05 (0.23)	0.07 (0.27)	0.04 (0.20)	0.29 (0.73)
Three-wall defect	0.10 (0.31)	0.16 (0.39)	0.08 (0.28)	0.07 (0.26)	0.16 (0.37)	0.14 (0.36)	0.20 (0.41)	0.36 (0.63)
Four-wall defect	0.07 (0.25)	0.09 (0.29)	0.08 (0.28)	0.07 (0.26)	0.11 (0.32)	0.14 (0.36)	0.08 (0.28)	0.07 (0.27)
PD (mm)	3.5 (0.6)	4.4 (0.8)	4.0 (0.7)	3.7 (0.7)	4.4 (0.7) **	4.4 (0.7) **	4.6 (0.8) ****	5.1 (0.8) ****##
Gingival recession (mm)	1.6 (1.1)	2.4 (1.1)	1.7 (1.0)	1.8 (1.1)	2.3 (0.7)	2.8 (1.0) *	2.9 (1.3) **#	3.0 (1.0) **#
CAL (mm)	2.1 (1.2)	3.7 (1.6)	2.5 (1.4)	2.4 (1.5)	3.7 (1.3) ***	4.2 (1.1) ****##	4.5 (1.4) ****####	5.1 (1.1) ****####
BOP (%)	19.9 (14.1)	47.6 (24.0)	29.8 (20.2)	36.7 (21.3)	51.7 (23.6) ****##	49.9 (21.4) ***#	58.2 (19.2) ****####	62.9 (23.4) ****####

Results are presented as mean (± standard deviation)

HCC Hepatocellular carcinoma, CNLC China Liver Cancer Staging, n Number, PD Periodontal probing depth, CAL Gingival recession, clinical attachment loss, BOP Bleeding on probing 

 * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 compared with control group

^# P < 0.05, ^## P < 0.01, ^### P < 0.001, ^#### P < 0.0001 compared with group Ⅰa

Comparisons of oral indicators between HCC patients and healthy volunteers

A full-mouth clinical examination was used to study the oral and periodontal health status of HCC patients at different stages (Table 1). The teeth and periodontal status of stage Ia and Ib were similar to those of control group. However, stage Ⅲa had more dental caries than the healthy (1.8 ± 1.6 vs. 4.1 ± 4.3, P < 0.05). Compared with the number of lost teeth of control group (0.9 ± 1.2), the number of lost teeth of stage Ⅱa (5.8 ± 6.0, P < 0.01), stage Ⅱb (5.8 ± 7.5, P < 0.05), stage Ⅲa (4.5 ± 5.5, P < 0.01), and stage Ⅲb (6.1 ± 6.9, P < 0.01) increased.

Regarding alveolar bone defects (AAP classification), no one-wall defects were found in any group. Two-wall defects were absent in the control group, but were only present in stage Ⅲb (0.29 ± 0.73). Three-wall defects were more prevalent in HCC patients (0.16 ± 0.39) than in controls (0.10 ± 0.31), with the highest value in stage Ⅲb (0.36 ± 0.63). Four-wall defects showed similar levels between controls (0.07 ± 0.25) and HCC patients (0.09 ± 0.29), though no statistically significant differences were noted for the above findings.

According to periodontal examination, we found that PD, CAL, BOP of stage Ⅱa, Ⅱb, Ⅲa, and Ⅲb were all significantly higher than control group (P < 0.01 for all comparisons). Similarly, gingival recession of stage Ⅱb (2.8 ± 1.0 mm, P < 0.05), stage Ⅲa (2.9 ± 1.3 mm, P < 0.01), and stage Ⅲb (3.0 ± 1.0 mm, P < 0.01) also exhibited an increase compared with control group (1.6 ± 1.1 mm). Notably, the most significant statistical differences in periodontal indicators were observed between the three-stage patients and control group (P < 0.0001 for all comparisons). Furthermore, we found that there were differences between HCC patients at different stages. Compared with PD, gingival recession, CAL, and BOP of stage Ia, all those indicators in stage Ⅲb exhibited an increase (5.1 ± 0.8 mm vs. 4.0 ± 0.7 mm, P < 0.01; 3.0 ± 1.0 mm vs. 1.7 ± 1.0 mm, P < 0.05; 5.1 ± 1.1 mm vs. 2.5 ± 1.4 mm, P < 0.0001; and 62.9 ± 23.4% vs. 29.8 ± 20.2%, P < 0.0001, respectively).

Because many HCC patients experienced dry and uncomfortable tongues, we also checked their tongues (Table 2). 23% of control group had tongue abnormalities, while in HCC patients, the proportion was as high as 57%, including fissured tongue (43%), atrophic glossitis (5%), swollen tongue papillae (4%), hairy tongue (3%), and ulcers (2%). The proportion of HCC patients with tongue diseases in stage Ia (41%) and stage Ib (28%) was lower than that in stage Ⅱa (70%), stage Ⅱb (50%), stage Ⅲa (70%), and stage Ⅲb (60%).

Table 2.

Symptoms of tongue in control group and HCC patients

Symptoms, n (%)	Control	HCC	CNLC stage of HCC
	n = 30	n = 113	Ⅰa	Ⅰb	Ⅱa	Ⅱb	Ⅲa	Ⅲb
			n = 24	n = 15	n = 19	n = 14	n = 26	n = 15
fissured tongue	6 (20%)	49 (43%)	8 (33%)	4 (27%)	13 (68%)	6 (43%)	10 (38%)	8 (53%)
atrophic glossitis	0 (0%)	6 (5%)	1 (4%)	0 (0%)	1 (5%)	0 (0%)	3 (12%)	1 (7%)
swollen tongue papillae	0 (0%)	5 (4%)	0 (0%)	0 (0%)	1 (5%)	1 (7%)	3 (12%)	0 (0%)
hairy tongue	0 (0%)	3 (3%)	1 (4%)	2 (1%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
ulcers	1 (3%)	2 (2%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	2 (8%)	0 (0%)

HCC Hepatocellular carcinoma, CNLC China Liver Cancer Staging, n Number

Correlations between periodontal indicators and liver-related clinical indicators

To investigate the correlation between periodontal indicators and liver-related clinical indicators in HCC patients, we collected their blood samples and conducted biochemical and coagulation tests at admission. Table 3 shows the characteristics of key liver-related clinical indicators in these patients. As shown in Table 4, correlation analysis demonstrated that PD was significantly correlated with AST (r = 0.63, 95% CI = 0.50–0.74, P < 0.0001), ALT (r = 0.31, 95% CI = 0.12–0.47, P = 0.0010), FIB (r = 0.70, 95% CI = 0.59–0.79, P < 0.0001), D-D (r = 0.65, 95% CI = 0.52–0.75, P < 0.0001), AFP (r = 0.57, 95% CI = 0.43–0.69, P < 0.0001), and PIVKA (r = 0.68, 95% CI = 0.57–0.77, P < 0.0001). And there was a weak correlation between PD and both TBIL (r = 0.19, 95% CI = -0.01-0.37, P = 0.0495) and IDBIL (r = 0.19, 95% CI = 0.00-0.37, P = 0.0480).

Table 3.

Characteristics of patients according to the different stages of HCC

	HCC	CNLC stage of HCC
	n = 113	Ⅰa	Ⅰb	Ⅱa	Ⅱb	Ⅲa	Ⅲb
		n = 24	n = 15	n = 19	n = 14	n = 26	n = 15
AST (U/L)	69.7 (158.8)	34.0 (16.9)	36.6 (17.4)	40.1 (34.0)	56.2 (32.7)	77.2 (54.0)	206.3 (423.3)
ALT (U/L)	44.5 (46.1)	40.2 (33.2)	33.8 (33.7)	31.1 (33.0)	42.4 (26.8)	49.4 (44.8)	74.6 (86.0)
ALP (U/L)	158.3 (118.3)	95.1 (34.4)	121.7 (42.2)	131.2 (121.8)	165.9 (99.8)	250.4 (168.5)	170.9 (69.76)
GGT (U/L)	149.2 (165.2)	52.6 (43.4)	100.2 (72.0)	115.9 (215.4)	123.3 (105.7)	267.2 (193.4)	216.6 (131.3)
CHE (U/L)	5329.0 (1685.0)	6302.0 (1667.0)	5303.0 (1188.0)	5417.0 (1376.0)	5025.0 (1966.0)	4611.0 (1886.0)	5154.0 (1286.0)
TBIL (umol/L)	21.9 (19.2)	17.8 (8.4)	16.8 (5.9)	19.5 (10.2)	21.9 (9.1)	27.6 (34.0)	27.2 (20.4)
DBIL (umol/L)	9.1 (13.4)	6.0 (3.4)	6.4 (2.9)	6.8 (6.4)	7.7 (3.9)	14.3 (24.2)	12.4 (15.5)
IDBIL (umol/L)	12.8 (7.6)	11.8 (6.1)	10.5 (4.5)	12.7 (6.0)	14.3 (6.2)	13.3 (11.3)	14.8 (7.7)
PT (s)	12.9 (1.4)	12.9 (1.2)	13.0 (1.6)	12.8 (1.4)	13.5 (1.4)	13.1 (1.4)	12.6 (1.1)
PTA (%)	82.9 (14.9)	83.1 (14.0)	86.1 (15.3)	80.7 (11.9)	74.8 (14.0)	82.7 (17.4)	88.6 (15.6)
APTT (s)	31.5 (4.2)	32.1 (4.2)	32.3 (4.2)	30.6 (3.3)	31.9 (3.1)	30.9 (4.1)	31.3 (5.9)
TT (s)	17.7 (2.5)	17.2 (3.6)	17.6 (1.7)	17.4 (1.7)	17.9 (1.7)	17.7 (1.5)	18.6 (3.6)
FIB (g/L)	3.0 (1.0)	2.4 (0.6)	2.9 (0.8)	2.7 (0.6)	2.6 (0.7)	3.9 (1.1)	3.3 (1.1)
D-D (mg/mL)	1.9 (4.5)	0.5 (0.3)	0.9 (0.8)	0.9 (0.9)	1.8 (2.7)	3.6 (6.5)	3.8 (7.8)
AFP (ng/mL)	2692.0 (8788.0)	282.6 (691.0)	2620 (5549.0)	697.2 (1489.0)	193.9 (204.3)	6669.0 (16779.0)	5001.0 (6089.0)
PIVKA (mAU/mL)	6862.0 (10811.0)	822.9 (2438.0)	7536.0 (10156.0)	6253.0 (11158.0)	11417.0 (12709.0)	7505.0 (11187.0)	11618.0 (13657.0)

Results are presented as mean (± standard deviation)

HCC Hepatocellular carcinoma, CNLC China Liver Cancer Staging, n Number, AST Aspartate transaminase, ALT Alanine transaminase, ALP Alkaline phosphatase, GGT: γ-glutamyl transpeptidase, CHE Cholinesterase, TBIL Total bilirubin, DBIL Direct bilirubin, IDBIL Indirect bilirubin, PT Prothrombin time, PTA Prothrombin time activity, APTT Activated partial thromboplastin time, TT Thrombin time, FIB Fibrinogen, D-D D-dimer, AFP Alpha-fetoprotein, PIVKA protein induced by vitamin K absence or antagonist-II

Table 4.

Correlation analysis of PD and laboratory indicators in HCC patients

	Spearman r	95% CI	P value
AST	0.63	(0.50, 0.74)	0.0001
ALT	0.31	(0.12, 0.47)	0.0010
ALP	0.14	(-0.05, 0.33)	0.1296
GGT	0.16	(-0.04, 0.35)	0.1001
CHE	-0.09	(-0.28, 0.10)	0.3502
TBIL	0.19	(-0.01, 0.37)	0.0495
DBIL	0.10	(-0.10, 0.28)	0.3071
IDBIL	0.19	(-0.00, 0.37)	0.0480
PT	-0.02	(-0.21, 0.17)	0.8136
PTA	-0.03	(-0.22, 0.17)	0.7843
APTT	-0.16	(-0.34, 0.04)	0.1105
TT	0.01	(-0.18, 0.20)	0.9132
FIB	0.70	(0.59, 0.79)	0.0001
D-D	0.65	(0.52, 0.75)	0.0001
AFP	0.57	(0.43, 0.69)	0.0001
PIVKA	0.68	(0.57, 0.77)	0.0001

PD Periodontal probing depth, HCC Hepatocellular carcinoma, CI Confidence interval, AST Aspartate transaminase, ALT Alanine transaminase, ALP Alkaline phosphatase, GGT γ-glutamyl transpeptidase, CHE Cholinesterase, TBIL Total bilirubin, DBIL Direct bilirubin, IDBIL Indirect bilirubin, PT Prothrombin time, PTA Prothrombin time activity, APTT Activated partial thromboplastin time, TT Thrombin time, FIB Fibrinogen, D-D D-dimer, AFP Alpha-fetoprotein, PIVKA Protein induced by vitamin K absence or antagonist-II

In addition, we assessed the link between liver-related clinical indicators and BOP, which is an indicator that reflects the severity of gingival inflammation (Table 5). Results indicated that BOP was significantly correlated with AST (r = 0.31, 95% CI = 0.12–0.47, P = 0.0010), ALP (r = 0.20, 95% CI = 0.01–0.38, P = 0.0347), GGT (r = 0.20, 95% CI = 0.00-0.38, P = 0.0409), APTT (r = 0.35, 95% CI = 0.17–0.51, P = 0.0002), and D-D (r = 0.30, 95% CI = 0.11–0.47, P = 0.0018).

Table 5.

Correlation analysis of BOP and laboratory indicators in HCC patients

	Spearman r	95% CI	P value
AST	0.31	(0.12, 0.47)	0.0010
ALT	-0.01	(-0.20, 0.18)	0.9283
ALP	0.20	(0.01, 0.38)	0.0347
GGT	0.20	(0.00, 0.38)	0.0409
CHE	-0.11	(-0.30, 0.08)	0.2341
TBIL	0.05	(-0.14, 0.24)	0.6096
DBIL	0.05	(-0.14, 0.24)	0.5899
IDBIL	-0.00	(-0.19, 0.19)	0.9901
PT	-0.01	(-0.20, 0.19)	0.9464
PTA	0.09	(-0.11, 0.28)	0.3530
APTT	0.35	(0.17, 0.51)	0.0002
TT	-0.07	(-0.26, 0.12)	0.4550
FIB	0.09	(-0.11, 0.28)	0.3473
D-D	0.30	(0.11, 0.47)	0.0018
AFP	0.07	(-0.13, 0.26)	0.4872
PIVKA	0.09	(-0.11, 0.27)	0.3601

BOP Bleeding on probing, HCC Hepatocellular carcinoma, CI Confidence interval, AST Aspartate transaminase, ALT Alanine transaminase, ALP Alkaline phosphatase, GGT γ-glutamyl transpeptidase, CHE Cholinesterase, TBIL Total bilirubin, DBIL Direct bilirubin, IDBIL Indirect bilirubin, PT Prothrombin time, PTA Prothrombin time activity, APTT Activated partial thromboplastin time, TT Thrombin time, FIB Fibrinogen, D-D D-dimer, AFP Alpha-fetoprotein, PIVKA Protein induced by vitamin K absence or antagonist-II

We applied ordinal logistic regression analysis to assess the association between oral health indicators and HCC stages (Table 6). HCC stages were defined as the ordinal dependent variable. Confounders including age, smoking history, alcohol intake and BMI were incorporated. The model passed the test of parallel lines (χ² = 27.080, P = 0.514). This verified the appropriateness of the ordinal logistic framework. The results showed consistent significant associations between oral health indicators and advanced HCC stages. Each millimeter increase in PD was linked to 181% higher odds of HCC stage progression (adjusted odds ratio [OR] = 2.81, 95% CI = 1.86–4.25; P < 0.001). BOP also had a significant association with elevated odds of advanced HCC (adjusted OR = 1.04, 95% CI = 1.03–1.05; P < 0.001). Similarly, the number of lost teeth was significantly related to more advanced HCC stages (adjusted OR = 1.16, 95% CI = 1.08–1.24; P < 0.001). Among confounders, smoking history remained significantly associated with HCC stage advancement (adjusted OR = 1.94, 95% CI = 1.02–3.73; P = 0.049). BMI (adjusted OR = 0.92, 95% CI = 0.84–1.01; P = 0.076), age (adjusted OR = 0.98, 95% CI: 0.95–1.01; P = 0.217) and alcohol intake (adjusted OR = 0.60, 95% CI = 0.30–1.21; P = 0.158) had no statistically significant links with HCC stages.

Table 6.

Ordinal logistic regression analysis of the association between oral indicators and stages of HCC

	Adjusted OR	95% CI	P value
BMI	0.92	(0.84, 1.01)	0.076
Age, years	0.98	(0.95, 1.01)	0.217
Smoking history	1.94	(1.02, 3.73)	0.049
Alcohol intake	0.60	(0.30, 1.21)	0.158
PD (mm)	2.81	(1.86, 4.25)	< 0.001
BOP (%)	1.04	(1.03, 1.05)	< 0.001
Lost teeth (n)	1.16	(1.08, 1.24)	< 0.001

HCC Hepatocellular carcinoma, OR Odds ratio, CI Confidence interval, BMI Body mass index, PD Periodontal probing depth, BOP Bleeding on probing, n number

To further explore the associations, we subsequently conducted multivariate ordinal logistic regression analyses with HCC stages as the independent variable, PD and laboratory indicators as the dependent variables, while adjusting for smoking as a confounder. Due to the strong correlation among various laboratory indicators, we analyzed each laboratory indicator separately; this approach ensured that each model passed the test of parallel lines, thereby validating the reliability of the ordinal logistic regression results (Table 7). Among laboratory indicators, several markers exhibited associations: ALT (adjusted OR = 1.01, 95% CI = 1.00–1.02; P = 0.043), IDBIL (adjusted OR = 1.01, 95% CI = 1.00–1.01; P = 0.004), and AFP (adjusted OR = 1.01, 95% CI = 1.00–1.01; P < 0.001) all had significant but modest effect magnitudes. In contrast, D-D showed a more meaningful laboratory-based association (adjusted OR = 1.67, 95% CI = 1.03–2.72; P = 0.038), with elevated D-D levels correlating with advanced HCC.

Table 7.

Ordinal logistic regression analysis of the association between laboratory indicators and stages of HCC

	Adjusted OR	95% CI	P value
Model 1
PD	2.07	(1.24, 3.43)	0.006
Smoking history	2.55	(1.30, 5.01)	0.008
ALT	1.01	(1.00, 1.02)	0.043
Model 2
PD	2.85	(1.80, 4.51)	<0.001
Smoking history	2.47	(1.20, 4.07)	0.014
IDBIL	1.01	(1.00, 1.01)	0.004
Model 3
PD	1.85	(1.04, 3.26)	0.035
Smoking history	2.50	(1.23, 5.08)	0.011
D-D	1.67	(1.03, 2.72)	0.038
Model 4
PD	2.77	(1.76, 4.37)	<0.001
Smoking history	2.08	(1.02, 4.27)	<0.001
AFP	1.01	(1.00, 1.01)	0.043

HCC Hepatocellular carcinoma, OR Odds ratio, CI Confidence interval, PD Periodontal probing depth, ALT Alanine transaminase, IDBIL Indirect bilirubin, D-D D-dimer, AFP Alpha-fetoprotein

Discussion

HCC ranks fourth among newly diagnosed cases of various cancers in China and represents the second leading cause of cancer-related mortality, thus posing a severe threat to the life and health of the Chinese population [11]. In the pathogenesis of HCC, diverse pathogenic factors trigger hepatocyte damage-repair cycles, leading to progressive liver fibrosis and structural destruction and eventually contributing to HCC occurrence [12]. Periodontitis, a known risk factor for multiple systemic diseases, can induce systemic responses such as inflammation and infection. Studies have found that periodontal disease and tooth loss are positively correlated with various liver abnormalities, including elevated transaminase levels, non-alcoholic fatty liver disease (NAFLD), cirrhosis, and liver cancer [13–15]. A study found that patients with severe periodontitis have a higher cirrhosis-related mortality rate [16]. What’s more, the risk of liver cancer in periodontitis patients was 1.32 times higher than that in patients without periodontitis [17].These findings emphasize the need to pay more attention to oral health in clinical care of HCC patients.

In the present study, we found an association between periodontal disease and HCC. Compared with stage Ia, stage Ⅲb showed increased PD, gingival recession, CAL, and BOP. Our results demonstrated that HCC patients had poorer oral health than healthy individuals, with this difference being more pronounced in patients with advanced HCC (stage III). This finding aligns with a case-control study [18], which reported that compared with non-periodontitis patients, periodontitis patients had poorer cancer staging and liver function, and thus a poorer prognosis. Notably, it has been found that cirrhotic patients have significantly poorer clinical periodontal parameters and a higher incidence of periodontitis than healthy individuals [19]. Our results in HCC patients are consistent with this finding, providing additional evidence for the study of the link between liver disease and periodontitis. Moreover, we found that HCC patients had a higher prevalence of tongue disorders, including fissured tongue, atrophic glossitis, swollen tongue papillae, hairy tongue, and ulcers. We found that dry mouth and mucosal lesions are also prevalent in people with liver disease other than HCC, which may be related to their own weakness and medication use. Systemic diseases and many medications may impair salivary flow. Dry mouth is a major risk factor for dental caries and predisposes to periodontitis and oral mucosal lesions, such as candidiasis, angular cheilitis, and painful stomatitis, all of which have been reported in CLDs [20]. A fissured tongue is common in dry mouth, and its prevalence in liver disease is around 30–40% [21]. And atrophic tongue is typical in alcohol-related liver disease [22]. Fissured tongue and atrophic tongue were significantly associated with hyposalivation in a study of liver transplant candidates [23].

Via regression analysis and multivariate ordinal logistic regression analyses, we verified the association between oral health status and HCC staging. After adjusting for confounding factors, increases in PD, BOP, and the number of missing teeth were significantly correlated with more advanced HCC stages. It was reported that individuals presenting 11–31 permanent teeth loss or all teeth lost at the baseline (surrogate markers of periodontitis) were more likely to develop HCC (hazard ratio = 1.42, 95% CI = 1.01–1.98, and hazard ratio = 1.45, 95% CI = 1.00-2.10) [24]. In the present study, the number of missing teeth recorded included those lost due to dental caries and periodontitis, while excluding teeth lost to trauma or congenital absence. Therefore, the findings regarding the correlation between missing teeth and HCC staging cannot be simply interpreted as an indicator specific to periodontitis. Nonetheless, this result still reflects the existence of an association between oral health status and HCC.

Among the confounding factors evaluated, smoking history remained significantly associated with HCC stage advancement. In contrast, BMI, age, and alcohol intake showed no statistically significant associations with HCC staging. This null effect of alcohol on the link between periodontitis and liver disease is consistent with findings from a prior study focusing on liver cirrhosis, which reported that the prevalence of periodontitis was twice as high in patients with cirrhosis compared with healthy controls, independent of alcohol use status [25].

We selected key HCC diagnostic indicators and liver function parameters for further investigation into their associations with PD and HCC staging. After adjusting for smoking as a confounding variable, D-D, ALT, DBIL, and AFP remained clinically relevant as potential auxiliary indicators for HCC staging in the setting of periodontitis. Collectively, these findings underscore the potential utility of oral health parameters in HCC disease stratification. The elevation of AST and ALT mainly reflects hepatocyte damage [26]. Hepatitis, cirrhosis, and other conditions can cause damage to hepatocytes, impairing bilirubin metabolism and excretion, which increases TBIL and IDBIL [27]. AFP and PIVKA are key HCC diagnostic markers [28]. Moreover, patients with elevated D-D are more likely to experience tumor recurrence and metastasis, and their risk of death are also higher. Therefore, D-D can serve as an important marker for monitoring the condition and evaluating the prognosis of patients with malignant tumors [29].

At present, research on the HCC-periodontitis correlation remains limited. However, other CLDs and periodontitis share multiple common risk factors. For example, NAFLD patients with hepatocyte mitochondrial overload exhibit increased the production of reactive oxygen species (ROS), and excessive ROS production is also a common phenomenon in periodontitis patients. Consequently, moderate-to-severe periodontitis increases the risk of liver fibrosis in NAFLD [30, 31]. Liver fibrosis can progress to more severe conditions, and notably, cirrhosis of unknown etiology is associated with periodontitis (OR = 2.28, 95% CI = 1.5–3.48) [14]. Additionally, HCC patients with severe periodontitis have higher levels of peripheral system ROS than those without periodontitis [18]. We proposed several speculations regarding these findings. Periodontitis may affect the levels of some liver injury factors in HCC patients. ALT, an important biomarker of liver injury, has been linked to increased PD in a cohort study [32]. Moreover, patients with cirrhosis and liver cancer are more susceptible to oral problems such as dental plaque and tartar, which exacerbate the elevation of serum ALP [33].

Given the close relationship between periodontitis and HCC identified herein, we need to give sufficient attention to the oral health of HCC patients. Oral microbes and their metabolites can act via the oral-liver axis impair mucosal immune function and trigger chronic excessive long-term immune responses, promoting liver diseases (e.g., fibrosis, cirrhosis, even HCC) [34]. Additionally, the ecological imbalance of the gut microbiota induced by periodontal bacterial colonization may disrupt the intestinal barrier, facilitating the translocation of liver injury factors and enterobacteria to the liver via enterohepatic circulation [35].

Given that we identified an association between PD, BOP, and the number of missing teeth and HCC, providing optimal oral treatment and periodontal maintenance may support HCC management. Specifically, untreated periodontitis in HCC patients can lead to adverse outcomes: chronic inflammation promotes the conversion of liver progenitor cells to liver cancer stem cells [36]. Moreover, bacteria, endotoxins (e.g., lipopolysaccharide), and locally produced inflammatory mediators can translocate into the bloodstream from inflamed periodontal tissues, inducing low-grade chronic systemic inflammation and impairing distant organs [20]. This risk is particularly notable for HCC patients undergoing surgeries (e.g., liver resection, liver transplantation), as oral infections may increase their systemic infection risk. Indeed, previous research has shown that liver transplant patients generally lack adequate preventive dental care, and when combined with immunosuppression, this further increases their susceptibility to bacterial and fungal infections [37, 38]. Fortunately, effective periodontal intervention can mitigate such risks: initial periodontal treatment and subgingival scaling with antibiotics have been shown to significantly reduce levels of systemic inflammatory markers, including C-reactive protein (CRP) and serum interleukin-6 (IL-6) [39, 40]. Thus, strengthening oral health management and promoting interdisciplinary collaboration to enhance oral health monitoring and guidance for patients pre- and post-surgery are essential.

This study analyzed the disparities in oral health status among HCC patients at different CNLC stages, confirming a correlation between periodontitis and HCC—specifically, advanced-stage HCC was associated with poorer periodontal conditions. This finding supplements the existing literature on periodontitis and HCC, which represents one of the major strengths of the present study. Nevertheless, several limitations should be acknowledged. Although we verified that PD, BOP, and tooth loss were correlated with HCC staging, we cannot directly determine the temporal sequence of the association: it remains unclear whether periodontitis precedes and contributes to HCC progression, or whether the deterioration of periodontal status in HCC patients is secondary to systemic immune dysfunction and nutritional impairment associated with the malignancy itself. Therefore, a long-term follow-up study is warranted in future research to evaluate the persistent impact of periodontitis on HCC. Furthermore, despite the potential benefits of active periodontal intervention and oral hygiene maintenance, the majority of patients included in this study declined such management strategies primarily due to psychological barriers (e.g., emotional distress related to cancer diagnosis) and financial constraints. This limits the generalizability of our findings regarding the efficacy of oral care interventions in this patient population. Nonetheless, based on a comprehensive review of published literature, we postulate that optimal oral care and periodontal treatment are undoubtedly beneficial for improving the quality of life and potentially the prognosis of HCC patients.

Conclusion

This study analyzed the oral health status of HCC patients at different CNLC stages, and clearly confirmed a significant correlation between periodontitis and HCC. Specifically, advanced-stage HCC patients were generally associated with increased PD, BOP, and a higher number of missing teeth. This result enriches the existing evidence on the association between periodontitis and HCC, and provides a new entry point for exploring the clinical management of HCC from the perspective of oral health. In summary, oral health status may serve as a potential reference indicator for assessing the condition and prognosis of HCC patients, and attaching importance to and strengthening oral health management in HCC patients is expected to become an important auxiliary means to improve their overall prognosis.

Abbreviations

AAP

American Academy of Periodontology

ALP

Alkaline phosphatase

ALT

Alanine transaminase

ANOVA

One-way analysis of variance

APTT

Activated partial thromboplastin time

AST

Aspartate transaminase

BMI

Body Mass Index

CAL

Clinical attachment loss

CBCT

Cone beam computed tomography

CEJ

Cementoenamel junction

CHE

Cholinesterase

CI

Confidence interval

CLD

chronic liver diseases

CNLC

China Liver Cancer Staging

DBIL

Direct bilirubin

D-D

D-dimer

DMFT

Decayed, missing, and filled teeth

FIB

Fibrinogen

GGT

γ-glutamyl transpeptidase

HCC

Hepatocellular carcinoma

IDBIL

Indirect bilirubin

NAFLD

Non-alcoholic fatty liver disease

OR

Odds ratio

PD

Periodontal probing depth

PIVKA

Protein induced by vitamin K absence or antagonist-II

PT

Prothrombin time

PTA

Prothrombin time activity

ROS

Reactive oxygen species

SD

Sstandard deviation

TBIL

Total bilirubin

TT

Thrombin time

Authors’ contributions

Data collection and assessment, writing, review, and editing: Liangqiuyue Zhong. Data collection and assessment: Wenjing Wang and Yunshan Zhao. Writing - review and editing: Chunshen Li, Pengyu Chen, and Yan Liu. Methodology: Xi Chen and Xiaogang Zhang. Funding acquisition: Xi Chen. Supervision and project management: Xi Chen and Xiaogang Zhang. All authors have read and agreed to the published version of the manuscript.

Funding

The study received support by Nanjing Tianqing Research Fund (No. HX202323).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted at the First Affiliated Hospital of Xi’an Jiaotong University in Xi’an, Shaanxi Province, China, in adherence to the ethical principles of the Declaration of Helsinki. The study protocol was approved by the hospital’s Institutional Review Board (No. XJTU1AF2024LSYY-057) prior to initiating data collection and analysis. Written informed consent was obtained from all participants before their enrollment in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.