This study included 20 Togolese patients diagnosed with chronic hepatitis B who were followed at the Hepato-gastro-enterology department, CHU Campus of Lomé and 20 healthy controls (HC) recruited at the National Blood Transfusion Center (CNTS) of Lomé from January to December 2025. Patients with chronic hepatitis B were defined by the following criteria: positive serum hepatitis B surface antigen (HBsAg) at least 6 months before obtaining blood samples [16] or positive serum HBsAg and anti-HBc IgG, anti-HBc IgM negative on one occasion [17]. None of the patients had received antiviral treatment or immunosuppressive therapy before sampling. A semi-structured questionnaire was used to obtain participants’ data including demographic and socioeconomic parameters, screening associated signs and symptoms, risk factors and treatment received. Ten mL venous blood samples were collected from each participant in an ethylene diamine tetraacetic acid (EDTA) tube and a tube without an additive.

2.1.1. Exclusion Criteria

For both groups, pregnant women, smokers, patients co-infected with other viruses (hepatitis C virus, hepatitis D virus, human immunodeficient virus), patients with autoimmune liver disease and liver cirrhosis, patients who had received antiviral therapy and patients complicated with other major organ diseases such as cardiovascular and renal diseases were excluded. For healthy controls, those who had abnormal total blood count, elevated liver enzymes such as alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were also excluded.

2.1.2. Definitions of Chronic HBV Infection Phases among HBeAg Negative Patients

Patients were classified into HBeAg negative chronic infection (HBeAg-infection) phase and HBeAg negative chronic hepatitis (HBeAg-disease) phase [11][14][18] after clinical investigations. HBeAg-infection phase is characterized by low (<2000 UI/mL) HBV DNA levels or 2000 UI/mL ≤ HBV DNA < 20,000 UI/mL with persistently normal alanine aminotransferase (ALT). However, HBeAg-disease phase is associated with HBV DNA > 2000 UI/mL and elevated ALT and/or other liver enzyme levels and/or clinical symptoms. According to viral load, they were also divided into active carrier or active replication group, defined as having HBV DNA > 2000 UI/mL and any ALT levels. Inactive carrier group is characterized by HBV DNA < 2000 UI/mL with ALT around the upper limit of normal (ULN), approximately 40 IU/L.

This study was approved by the bioethics committee for health research of Togo (“Comité de Bioéthique pour la Recherche en Santé”, CBRS) N˚ 039/2024/CBRS and all participants provided written informed consent before taking part in the study.

The biochemical indicators of liver enzymes, such as aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), and gamma-glutamyl transferase (GGT) were tested in serum samples using an automatic biochemical analyzer cobas c311 (Roche Diagnostics GmbH). The reference values described by Ribeiro et al. were used for parameter interpretations [19]. HBV serological markers as antibody to HBcAg (anti‐HBc), HBeAg, and antibody to HBeAg (anti‐HBe), and alfa-fetoprotein, were detected by the electrochemiluminescence assay (ECLIA) via the Cobas e411 platform (Elecsys; e411 Cobas; Roche Diagnostics Gmb). The white blood cells (WBC) and lymphocyte (LYM) counts were detected by an automatic five-classification blood cell analyzer (Maindray BC-6800, Shenzhen Mindray, China). Viral DNA was extracted according to GXT NA Extraction kit procedures on the automated instrument GenoXtract^® (Hain Lifescience, Nerhen, Germany) for plasma samples. To detect and quantify HBV DNA, quantitative real-time polymerase chain reaction (qPCR) was performed with GENERIC HBV CHARGE VIRALE (GHBV-CV) kit (BIOCENTRIC, Bandol, France) using FluoroCycler^®XT (Hain Lifescience, Nerhen, Germany) according to the manufacturer’s instructions.

2.4.1. Whole Blood Stimulation

Whole blood (WB) was stimulated in duplicate using 96-wells U-bottom plates (Greiner Bio One, Frickenhausen, Germany) as has been shown before [20]. In brief, 100 µL per well of WB sample were mixed with equal volume of 1X cell stimulation cocktail (PMA, Ionomycin, of Brefeldin A and Monensin) (eBioscience) and RPMI-1640 medium. Then plates were incubated at 37˚C in presence of 5% CO₂ for 6 hours after which red blood cells were lysed from using RBC Lysis Buffer (Roche^® Diagnostics GmbH, Mannheim, Germany) according to the manufacturer’s instructions.

2.4.2. Panel Configuration

Two separate panels were devised for acquisition to assess CD8⁺ T-cells functionality. The following fluorochrome-conjugated monoclonal antibodies were used: Panel 1: (CD8-APC, IFNγ-FITC, perforin-PE); Panel 2: (CD8-APC, IL17A-FITC, granzymeB-PE).

2.4.3. Cellular Immunostaining and Phenotyping

After red blood cells lysis, cell-surface staining was performed using CD8-APC (clone SK1) and incubated at 4˚C for 30 mins. Subsequently, cells were washed and permeabilized using Fix-Perm reagent (Invitrogen/Thermo Fisher Scientific) and Fc Receptor (FCR) blocking solution (human TruStainFcX^®, BioLegend, San diego, CA, USA) was used according to the manufacturer’s instructions to reduce unspecific antibody biding. For intracellular staining, cells were incubated at 4˚C with panel 1) Anti-human IFNγ-FITC (clone 4S.B3) and perforin-PE (clone dG9); panel 2) Anti-human IL17A-FITC (clone BL168) and granzymeB-PE (REF GRB04. After 30 minutes at 4˚C, cells were washed and resuspended in fix-perm buffer for acquisition. In order to test what portion of the cells were dead, although a whole blood sample was used without pretreatment and without additives that could impact cell viability, Zombie Aqua^™ Fixable Viability Kit (BioLegend) was included. Data were acquired and analyzed using a flow cytometer (Cytoflex, Beckman Coulter, Brea, California, USA) with CytExpert 2.1 software (Cytoflex, Beckman Coulter, Brea, California, USA). Instruments were calibrated using VersaComp Antibody Capture Bead Kit (Beckman Coulter, CA, USA) and CytoFLEX Daily QC Fluorospheres (Beckman Coulter, CA, USA), respectively, for compensation controls and for daily verification of the flow cytometry optical alignment and fluidics system. Fluorescence Minus One (FMO) controls were customized for each panel, including overlapping markers and unstained cells served as a negative control. The flow cytometry gating strategies are shown in Figure 1.

First of all, we searched a quality of run gate based on the time and the SSC-A surface area (A), then lymphocytes gate (total T cells) (B) were set based on cells granularity and size defined respectively by side scatter (SSC) and Forward scatter (FSC). From them, we evaluated dead and living cells proportions (C) before identified CD8⁺ T-cells (D) which were then separated into CD8⁺IL-17A⁺ (E); CD8⁺IFNγ⁺ (F); CD8⁺Perforin⁺ (G) and CD8⁺Granzyme B⁺ (H).

Data were collected using Excel and statistical analyses were performed with GraphPad Prism version 8.0 (GraphPad Software, San Diego California, USA) and SPSS version 25.0 manufactured by International Business Management (IBM SPSS Statistics version 25; Armonk, NY). The normality of data distribution was assessed by the Kolmogorov-Smirnov test. As the data sets were non-parametric, differences between the two groups were evaluated with non-parametric Mann-Whitney U tests. Correlation analysis was carried out by Spearman’s rank sum correlation. The association between the frequency of total CD8⁺ T cells, Tc1, Tc17 cells, as well as granule expressions with HBeAg negative disease and HBV replication prognosis, was analyzed in binary logistic regressions. Variables with a p-value < 0.20 in the bivariable analyses were subsequently incorporated into the multivariable logistic regression model. Fisher’s exact test was used to compare gender frequency. In all analyses, p-value < 0.05 was considered significant.

Figure 1. Gating strategies of flow cytometry phenotyping of CD8⁺ T cells.

We included forty individuals in this study, including 20 chronic Hepatitis B patients (CHB) and 20 healthy controls (HC). Demographic, virological and clinical characteristics of the enrolled subjects were summarized in in Table 1. All patients were positive for anti-HBe, anti-HBc (IgG) and negative for HBeAg, most of them never received a blood transfusion before their infection. The median age of chronic hepatitis B infection patients was 32.50 years and 60% were female, while the healthy controls had a median age of 29.50 years. The patients in HBeAg negative infection phase were the most represented (80%), showing a significant decrease in serum ALT (p = 0.0029), AST (p = 0.0046), and HBV DNA (p = 0.0020) levels compared with those with HBeAg negative hepatitis (20%). Classification of subjects based on viral load showed no statistically significant difference in liver enzyme levels as ALT, AST and others but however, a significant difference was noted in HBV DNA (p = 0.0002) levels (Table 2).

Table 1. Demographic, virological and clinical characteristics of the study population.

NA: not applicable, ALT: alanine aminotransferase; Reference values in chronic hepatitis: <31 U/L (women) and 37 U/L (men); AST: aspartate aminotransferase; Reference values in chronic hepatitis: <31 U/L (women) and <41 U/L (men); GGT: gamma-glutamyl transferase; Reference values 8 - 61 U/L (men) and 5 - 36 U/L (women); ALP: alkaline phosphatase; Reference values 40 - 150 U/L; AFP: alfa-fetoprotein; Reference values ≤ 5 U/L; WBC: white blood cells. HBsAg: hepatitis B surface antigen; HBeAg (−ve): hepatitis B e-antigen negative; Anti-HBc (+ve): hepatitis B core antibody positive; Anti-HBe (+ve): hepatitis B e-antibody positive; HBV DNA: hepatitis B virus desoxyribose nucleic acid.

Table 2. Comparison of clinical indexes at different phases of chronic hepatitis B.

NA: not applicable, ALT: alanine aminotransferase; Reference values in chronic hepatitis: <31 U/L (women) and 37 U/L (men); AST: aspartate aminotransferase; Reference values in chronic hepatitis: <31 U/L (women) and <41 U/L (men); GGT: gamma-glutamyl transferase; Reference values 8 - 61 U/L (men) and 5 - 36 U/L (women); ALP: alkaline phosphatase; Reference values 40 - 150 U/L; AFP: alfa-fetoprotein; Reference values ≤ 5 U/L; WBC: white blood cells; e-CI: HBeAg negative infection; e-CHB: HBeAg negative disease; AC: Active carrier; IC: Inactive carrier.

Cytotoxic T lymphocytes (CD8⁺ T cells) and their subsets were characterized in blood from both healthy control and chronic hepatitis B infection groups. We observed that compared to the healthy control group, the total percentage of CD8⁺ T cells was significantly decreased (p < 0.0001) in chronically infected subjects (Figure 2(A)). We also found that CHB subjects exhibited higher frequencies of CD8⁺ T lymphocytes that expressed IFNγ (Tc1) (Figure 2(B)) and IL-17A (Tc17), although this was only significant for CD8⁺IL17A⁺ T cells (p = 0.0029) (Figure 2(C)). Interestingly, both the CD8⁺IFN-γ⁺(p = 0.0153) and CD8⁺IL17A⁺(p = 0.0227) mean fluorescence intensity (MFI) were higher in this subject (Figure 3).

Data are expressed as median with interquartile range. Each dot represents an individual data point, and the horizontal black lines represent the median. CHB: chronic hepatitis B group (N = 20); HC: healthy control group (N = 20); The number of asterisks in the figures indicates the level of statistical significance (****p < 0.0001; **p < 0.01).

Figure 2. The percentage of circulating CD8⁺ T cells and their subsets in CHB patients and healthy controls.

Data are expressed as median with interquartile range of the mean fluorescence intensity of IFN-γ (A) and IL-17A (B). CHB: chronic hepatitis B group (N = 20); HC: healthy control group (N = 20); (*p < 0.05).

Figure 3. Comparison of CD8⁺IFN-γ⁺ and CD8⁺IL17A⁺ T cells mean fluorescence intensity (MFI) between CHB and HC.

The evaluation of CD8⁺ T cells and their subsets was carried out at different stages of chronic hepatitis and according to viral activity. No significant difference was found in CD8⁺ T cells percentage at different stages, whether in HBeAg negative phases (Figure 4(A)) or in the viral load phases (Figure 4(D)), nor in the percentage of CD8⁺IFNγ⁺ T cells (Figure 4(B) and Figure 4(E)) and even CD8⁺IL17A⁺T cells (Figure 4(C) and Figure 4(F)).

Data are expressed as median with interquartile range. Each dot represents an individual data point, and the horizontal black lines represent the median. Comparison of CD8⁺ T cells according to HBeAg negative phases (A) and viral load phases (B); Comparison of CD8⁺ T producing IFNγ (Tc1) at different stages of chronic hepatitis B infection according to HBeAg negative (C) and viral load (D); Comparison of CD8⁺IL-17A⁺ T cells at different stages of chronic hepatitis B infection according to HBeAg negative (E) and according to viral load (F). HBeAg-infection (N = 16); HBeAg-disease (N = 4); inactive carrier (N = 13); active carrier (N = 7).

Figure 4. Comparison of CD8⁺ T cells and it subsets percentage at different stages of chronic hepatitis B.

Considering that CD8⁺ T cells functionality is associated with progression to chronic hepatitis B, we then focused our analyses on their local function corresponding to the expression of cytotoxic molecules as perforin and granzyme B in the blood. The results showed that the frequencies of CD8⁺Perforin⁺ (Figure 5(A)) and CD8⁺GzmB⁺ (Figure 5(B)) T cells were not altered among CHB group compared to healthy controls. Similarly, no difference in Perforin (Figure 5(C)) and granzyme B (Figure 5(D)) expression by CD8⁺ T cells at HBeAg negative phases as well as at different stages according to viral load (Figure 5(E) and Figure 5(F)).

Comparison of CD8⁺perforin⁺ T cells (A) and CD8⁺granzymeB⁺ T cells (B)between chronic hepatitis B infection (CHB) group (N = 20) and HC, healthy control (N = 20). Perforin [(C),(E)] and Granzyme B [(D),(F)] expression by CD8⁺ T cells was also compared at different phases of chronic hepatitis B. The data are expressed as median with interquartile range.Each dot represents an individual data point, and the horizontal black lines represent the median. Statistical analysis was performed using a Mann-Whitney test. HBeAg-infection (N = 16); HBeAg-disease (N = 4); inactive carrier (N = 13); active carrier (N = 7).

Figure 5. Cytotoxic function of circulating CD8⁺ T cells from chronic hepatitis B patients.

The correlation of six clinical indexes, namely HBV DNA, ALT, AST, GGT, AFP, ALP levels on one hand and the expression of perforin and granzyme B by CD8⁺ lymphocytes on the other hand, with total CD8⁺ T cells, CD8⁺IFNγ⁺ and CD8⁺IL17A⁺ frequency was analyzed using the Spearman rank test. Alkaline phosphatase (ALP) serum level was negatively correlated with CD8⁺ T cells frequency (r = −0.678, p = 0.0010) in HBeAg negative CHB. In contrast, all five clinical parameters showed no correlation with CD8⁺ T cell subset frequencies (Table 3). A positive correlation was observed regarding perforin (r = 0.582, p = 0.007) expression with CD8⁺IFNγ⁺ and CD8⁺IL17A⁺ frequency.

Table 3. Correlation of ALT, DNA HVB, ALP and GGT serum levels, % CD8⁺Perf⁺ and % CD8⁺GrzmB⁺ with CD8⁺ T cells subsets frequencies in patients with CHB.

ALT, alanine aminotransferase; HBV DNA, hepatitis B virus desoxyribose nucleic acid; ALP: alkaline phosphatase; GGT: gamma-glutamyl transferase. Spearman’s correlation rank test was used for analyses.

Binary logistic regression analysis was performed to search for possible association between immunological parameters (total CD8⁺ T cells and their subpopulations) and Chronic Hepatitis B. Univariate analysis indicated a significant association between total CD8⁺ T cells (p = 0.001, odd ratio (OR) = 0.909; 80% confidence interval (CI) = 0.877 - 0.942), Tc17 (p = 0.017, OR = 1.176; 80% CI = 1.078 - 1.282), and CHB infection (Table 4). In multivariate analysis, total CD8⁺ T cells (aOR = 0.879, p = 0.002; 95% CI = 0.811 - 0.953) and Tc17 (aOR = 1.302, p = 0.008; 95% CI = 1.072 - 1.582) remained significantly associated with CHB. Among the 20 CHB patients included in our study, 7 showed viral replication (active carrier) and 4 showed symptoms and or disease (HBeAg negative disease). The correlation between the frequency of total CD8⁺, CD8⁺IFNγ⁺, CD8⁺IL17A⁺ T cells and granules expressions with HBeAg negative disease and viral load prognosis was also analyzed using binary logistic regression (Table 5). The results revealed that only CD8⁺IL17A⁺ T cells frequency was independent risk factor for HBeAg negative disease (p = 0.027, OR = 1.207, 80% CI = 1.022 - 1.427) and for HVB replication (p = 0.028, OR = 1.184, 80% CI = 1.018 - 1.378).

Table 4. Logistic regressions between CHB infection and CD8⁺ T cells response.

Tc1 = CD8⁺IFNγ⁺T cells; Tc17 = CD8⁺IL17A⁺ T cells; OR: odds ratio; CI: confidence interval.

Table 5. Binary logistic regression analysis of HBeAg negative disease and replication risk factors.

^aDependent variables, Tc1 = CD8⁺IFNγ⁺T cells ; Tc17 = CD8⁺IL17A⁺ T cells; OR: odds ratio; CI: confidence interval.

In summary, in patients with chronic hepatitis B, a decrease in circulating CD8⁺ T lymphocytes and an increase in CD8⁺IL17A⁺ T lymphocytes were observed compared to healthy controls. This elevated CD8⁺IL17A⁺ T lymphocytes was associated with chronic hepatitis B.