Impaired function of CD4+ T follicular helper (Tfh) cells associated with hepatocellular carcinoma progression

Abstract

Background and aims: CD4+ T follicular helper (Tfh) cells, a new subset of immune cells, have been demonstrated to be involved in the development and prognosis of tumors. However, their functional role in human hepatocellular carcinoma (HCC) is relatively unknown, and the detailed mechanisms in HCC development remain to be described.

Methods: A total of 85 HCC patients with hepatitis B virus (HBV) infection, 25 HBV-relative liver cirrhosis (LC) patients, and 20 healthy controls (HC) were randomly enrolled. Flow cytometric analysis, immunohistochemical staining, and relative function (i.e., cytokine secretion, B cell maturation) assays were used to analyze the properties of CXCR5+CD4+ T cells. In addition, the relationship between the frequency of CXCR5+CD4+ T cells and overall survival rates or disease-free survival rates was also analyzed by the Kaplan-Meier method.

Results: The frequency of circulating CXCR5+CD4+ T cells was significantly decreased in HCC patients compared with HBV-relative liver cirrhosis (LC) patients and healthy controls, and the decrease in circulating CXCR5+CD4+ T cells correlated with disease progression. The proportion of infiltrated CXCR5+CD4+ T cells was significantly decreased in tumor regions compared with nontumor regions. Furthermore, compared with healthy controls, the function of circulating CXCR5+CD4+ T cells in HCC was impaired, with reduced IL-21 secretion and dysfunction in promoting B cell maturation. Importantly, follow-up data indicated that a decreased frequency of circulating CXCR5+CD4+ T cells was also associated with reduced disease-free survival time in HCC patients.

Conclusions: Impairment of CD4+ T follicular helper cells may influence the development of HBV-associated HCC. Decreased CD4+ T follicular helper cells may represent a potential prognostic marker and serve as a novel therapeutic target for HCC patients.

Fig 1. Circulating CXCR5+CD4+ Tfh cells are significantly decreased in HCC patients.

(A) Representative prevalence of CXCR5+CD4+ Tfh cells from individual subjects in 3 studied groups and statistical analysis showing that the frequency of CXCR5+CD4+ Tfh cells in HCC patients is significantly lower than in HBV-LC patients and healthy controls. Each dot represents one individual. (B and C) Representative prevalence of ICOS^highCXCR5+CD4+ and PD-1^highCXCR5+CD4+ Tfh cells from individual subjects in 3 studied groups, significant correlations are found between the frequency of CXCR5+CD4+ Tfh cells and ICOS^high, PD-1^high. Each dot represents one individual. The solid line represents the linear growth trend. r, correlative coefficient. p values are shown.

Fig 2. A decrease in circulating CXCR5+CD4+ Tfh cells is associated with HCC progression, and CXCR5+CD4+ Tfh cells are decreased in tumor regions.

(A) The prevalence of CXCR5+CD4+ Tfh cells decreases with progressive stages in HCC. Each dot represents one individual. p values are shown. (B) Comparison of TILs with NILs shows that the frequency of CXCR5+CD4+ Tfh cells decreased significantly in the former. (C) Representative immunohistochemical staining of samples from HCC patients. Tfh cells are double stained for CD3 (blue, on cell membrane) and CXCR5 (red, on cell membrane). Quantitative analysis of Tfh-positive dots in tumor region and nontumor region from HCC patients. The data are collected at 200× power by two different investigators. One dot indicates one section from one patients. T, tumor region; nT, nontumor region.

Fig 3. Correlations between the frequency of CXCR5+CD4+ Tfh cells and clinical parameters.

The solid line represents the linear growth trend. r, correlative coefficient. p values are shown. AFP, α-fetoprotein; AST, aspartate aminotransferase; ALT, alanine aminotransferase.

Fig 4. Tfh cells from HCC patients secrete significantly lower levels of IL-21 and IFN-γ than Tfh cells from HBV-LC individuals and healthy controls.

(A) Representative dot plots depict IL-21, IFN-γ, IL-17, and IL-10 expression on peripheral cells among the 3 groups in response to PMA/Iono stimulation. (B) Pooled data showing that the percentages of IL-21 and IFN-γ secreted by Tfh cells in HCC patients are significantly lower than in HBV-LC patients and healthy controls. There are no significant differences regarding IL-17 and IL-10 secreted by Tfh cells between the HCC and HBV-LC patients and healthy controls. Each dot represents one individual. p values are shown.

Fig 5. Tfh cells from HCC patients suppress the maturation of autologous B cells.

(A) Circulating CXCR5+CD4+ Tfh cells from HCC patients (pTfh) or healthy controls (hTfh) and allogeneic CD19+ B cells from healthy controls (hB) are incubated at a ratio of 1:1 in the presence of SEB (100 ng/ml) in RPMI1640 complete medium; after 7–8 days of culture, the B cell subpopulations are examined by the surface phenotype. The data are a representative HCC patient and healthy control. (B) Statistical analysis showing that class-switched memory B cells are significantly decreased in the co-culture systems of pTfh and allogeneic hB compared with hTfh and hB. The data are from 10 HCC individuals and 10 healthy controls.

Fig 6. Decreased prevalence of circulating CXCR5+CD4+ Tfh cells predicts a poor survival in HCC.

(A) 43 HCC patients in early stages of the disease (BCLC 0, A, B) are divided into two groups according to the median value of peripheral Tfh cells frequency (median value for BCLC 0, A and B, 9.4%). The low-Tfh group (n = 22, Tfh frequency: 6.7% ± 1.8%) has significantly poorer disease-free survival (DFS) rates compared to the high-Tfh group (n = 21, Tfh frequency: 11.8% ± 1.8%) (p = 0.003). (B) The other 42 HCC patients with advanced disease (BCLC C, D) are also divided into two groups according to the median value of peripheral Tfh cells frequency (median value for BCLC C and D, 7.4%). However, the circulating Tfh cells don’t have any significant predictive value (5.4% ± 1.5% vs 10.1% ± 3.1%, p = 0.159) for overall survival (OS) rates. The DFS and OS rates are analyzed by the Kaplan-Meier method, and the long-rank test is applied between-group comparisons.