Background
Gastric cancer (GC), one of the leading causes for worldwide mortality, is considered as the second most common cancer in China [
1,
2]. Although the clinical management for GC is diverse, the prognosis of GC remains poor [
3]. The mechanism of clinical management is supposed to include increasing regulatory T-cells (Tregs), losing tumor antigen expression, and enhancing tumor expression of inhibitory ligand [
4‐
7]. Reports proved that the depletion of Tregs could not enhance the efficacy of primary therapies in some cancers, which suggests that T cells may work in synergy with other mechanisms to suppress anti-tumor immunity [
8‐
10].
Investigations showed that T-cell dysfunction in chronic virus infection and human tumor growth was related to the up-regulation of inhibitory molecules such as programmed death 1 (PD-1), T-cell immunoglobulin, and mucin-domain-containing molecule 3 (Tim-3) [
11‐
14]. It provides a significant scheme for cancer treatment that invalidates these inhibitory pathways to resume exhausted T-cells. However, only 19.5% of GC patients respond to PD-1 inhibition [
15].
Currently, there is little detail of T-cell subsets resident within gastric cancer tissues or the expression patterns in this microenvironment corresponding to those observed in paraneoplastic tissue or in the peripheral blood of GC patients. Moreover, the role of PD-1+, Tim-3+, and Tregs in the development and maintenance of tumor-infiltrating T-cell dysfunction in GC patients need to be investigated. Thus, it is crucial to discover the complex mechanisms for inducing T-cell dysfunction in GC patients by exploring the subset composition and functional properties of tumor-infiltrating T-cells.
In this study, we investigated the distribution of T-cells subset, the differentiation and inhibitory phenotype of T-cells from blood and tissues of GC patients. The results demonstrated that the T-cell subsets resident within GC is different from that in paraneoplastic tissue and blood, and the increase in Tim-3+ PD-1+ CD4+ T-cells in tumor tissues was found to correlate with the clinical cancer stage and Tregs. These data as well as the corresponding results and the methods provide a new immunotherapeutic approach for the clinical management of GC.
Materials and methods
Patients
Patients with GC diagnosed on the basis of pre-operative staging and laparotomy findings were approached for enrollment between 2014 and 2015 at the Department of Surgery, the Affiliated Cancer Hospital of Zhengzhou University, China. Thirty-one patients diagnosed with primary GC without previously treatment were qualified in this study. Peripheral blood samples were collected from each patient before treatment. The clinical stage was classified according to the American Joint Committee on Cancer (AJCC) Staging Manual, Seventh Edition (2010). The clinicopathological characteristics of age, gender, and histological grade at the time of blood sample collection were recorded in Table
1.
Table 1
Characteristics of patients
Total patients (n) | | 31 |
Median age in years (range) | | 60 (37–79) |
Gender | Male | 21 |
| Female | 10 |
AJCC stage (2010) | I/II | 12 |
| III | 19 |
Histologic differentiation | Well/moderate | 13 |
| Poor | 18 |
Weight loss (%) | None | 4 |
| ≤ 10 | 16 |
| > 10 | 11 |
Smoking status | Current | 17 |
| Ex-smoker | 4 |
| Never | 10 |
Phenotypic analysis
Fresh venous blood was collected from patients or healthy donors (HDs) with EDTA-coated vacutainer tubes. Specific anti-CD3-PE-Cy7 (BioLegend, San Diego, CA, USA) or anti-CD3-FITC (BioLegend, San Diego, CA, USA), anti-CD25-PE (BD Pharmingen™, San Diego, California, USA), anti-CD127-APC (BioLegend, San Diego, CA, USA), anti-CD4-FITC (BioLegend, San Diego, CA, USA), anti-CD8-APC (BioLegend, San Diego, CA, USA), anti-CD45RA-FITC (BioLegend, San Diego, CA, USA), anti-CCR7-PE-Cy7 (BioLegend, San Diego, CA, USA), anti-CD4-PE (BD Biosciences, Oxnard, CA), anti-PD-1-FITC (BioLegend, San Diego, CA, USA), anti-CD4-PerCP-cy5.5 (BD Biosciences, Oxnard, CA), anti-Tim-3-PE (BD Pharmingen™, San Diego, California, USA), and anti-IFN-γ-APC (eBioscience, San Diego, CA, USA) were purchased for use. The viability of the cells was assessed by a violet amine reactive dye (Invitrogen, Carlsbad, CA).
Briefly, 50 μL of blood mixed with 5 μL of each antibody was incubated on ice for 20 min in the dark. Then, 2 mL of 1× lysis buffer (BD Biosciences, Oxnard, CA) was added to each sample and incubated at room temperature for 15 min. Samples were washed with FACS buffer (5% BSA in PBS, 0.09% sodium azide), and the pellets were resuspended in 300 μL of FACS buffer. Flow cytometry was performed on a BD FACS Aria II flow cytometer, and was analyzed with FlowJo software (TreeStar Inc., Ashland, OR, USA). The values were background-corrected by control sample or fluorescence minus one (FMO). Cells without surface makers were used as control samples.
Isolation of TILs
The fresh human tumor samples and matched paraneoplastic tissues from patients with GC were cut into pieces (3 ~ 5 mm3) and were treated with 1 μg/mL of collagenase (Sigma-Aldrich, St. Louis, MO, USA), 25 μg/mL of DNase (Sigma-Aldrich, St. Louis, MO, USA), and 2% fetal bovine serum in PBS at 37 °C for 1 to 1.5 h. The tissue homogenates were filtered by a 70-μm cell strainer (Falcon; BD Biosciences, Oxnard, CA) before density centrifugation. Density centrifugation was performed using Percoll density gradient, and cells at the interface between 40 and 80% discontinuous Percoll gradient were collected. The leukocyte’s viability was evaluated by trypan blue exclusion.
Intracellular IFN-γ staining assay
To evaluate the effect of the blockade of PD-1/PD-L1/2 and Tim-3/Tim-3-L pathways on the induction of IFN-γ, TILs and T-cells in noncancerous tissues were pre-incubation with blocking antibodies for 1 h, and stimulated with 5 μg/mL of anti-CD3 (eBioscience, San Diego, CA, USA) for 6 h [
14]. After incubation for another 2 h, 10 μg/mL of Brefeldin A (Sigma- Aldrich, St. Louis, MO, USA) was added to the culture medium. Cells were then stained with antibodies anti-CD3-PE-Cy7 and anti-CD4-PerCp-Cy5.5/anti-CD8-PerCp-Cy5.5, and marked with mAb against IFN-γ-APC (eBioscience, San Diego, CA, USA). Five hundred thousand events were recorded during flow cytometric analysis.
Statistical analysis
Statistical analysis was performed by GraphPad Prism 5.0 (GraphPad Software, US). Mann–Whitney test was used to assess the differences between the study groups. A pair wised t test was applied to compare the expression of the inhibitory molecules in cancer, noncancerous tissues, and blood. p < 0.05 was considered to be statistically significant.
Discussion
It has been reported that T lymphocytes played a critical role in controlling and eliminating cancer [
3,
17,
18]. Researches revealed that tumor antigen would activate cytotoxic CD8
+ T-cells which was enhanced by specific helper CD4
+ T-cells [
19,
20]. In our study, the GC tumors have higher percentages of CD4
+ T-cells and a lower frequency of CD8
+ T-cells compared with that in paraneoplastic tissues or blood, which indicated an increased CD4
+/CD8
+ ratio in tumors. In addition, the CD4
+/CD8
+ ratio in peripheral blood was significantly higher than that in matched paraneoplastic tissue. Our results showed that the T-cell subset distribution in patients with GC was different for varied tissues. The comparison is essential to the rational design for novel immuno-therapeutics strategy against GC. Previous reports showed that CD8
+ T-cells prevailed over CD4
+ T-cells in the tumor lesions derived from neuroblastoma patients, which meant that the CD4
+/CD8
+ ratio in tumors was lower than that in peripheral blood [
20]. However, these results were quite different from ours. In addition, the percentage of CD4
+ T-cells was decreased in livers when compared with the blood in inflammatory liver injuries and viral infections [
21]. According to these studies, we conclude that different molecular mechanisms may cause the skewing of the CD4
+/CD8
+ T-cell ratio and the reason is not clear for the changes of CD4
+ T-cells in GC patients, which needs further clinical investigations.
Previous studies have reported that there were increased frequency of effector and memory T-cells among virus-specific CD4
+ T-cells during acute-resolving and chronic viral infections [
18,
19]. In our research, the frequency of effector and memory T-cells among TILs and paraneoplastic T-cells elevated comparing with that of circulating T-cells, which was coincided with those in neuroblastoma patients. Fridman et al. found that the presence of CD8
+ memory T-cells was associated with a favorable prognosis [
22]. However, the relationship between the presence of CD4
+ memory T-cells and prognosis in patients with GC has yet to be elucidated, and it will be explored in further researches.
In addition, regulatory T-cells can regulate antigens and suppress immunity to cancer [
10]. However, the role of Tregs in anticancer immunity is more controversial. In some cancers such as ovarian carcinoma and HPV-associated patients, high numbers of tumor-infiltrating Tregs have shown poor outcomes [
23,
24]. However, in colorectal cancer and lymphomas, Tregs have been reported as a positive prognostic factor [
25,
26]. These findings implied that the role of Tregs might vary according to the type and etiology of the cancer. In GC, Choi et al. demonstrated that high level of Tregs among tumor-infiltrating CD4
+ T-cells were favorable, but they only analyzed the Tregs in tumor tissue and peripheral blood from a healthy control [
3]. On the contrary, Hennequin et al. showed that low infiltration of Tregs were associated with better relapse-free survival in patients with localized gastric cancer [
11]. Moreover, there is still no definitive markers that used to define Treg. Instead, CD127
low expression, which is known to be highly enriched in regulatory CD4
+ CD25
+ T-cells, might be the most specific marker that is known and used so far [
27‐
29]. Therefore CD4
+ CD25
high CD127
low was used to assess Tregs in CD4
+ T-cells of GC patients in our study. We discovered an elevated level of Tregs in peripheral blood from patients with GC when compared with those from HDs. Notably, the percentage of CD25
high CD127
low T regulatory cells among TILs were significantly higher than their counterparts in peripheral blood and paraneoplastic tissue, while the Tregs in peripheral blood has no significant difference as compared with that in Nils, which indicated that the GC milieu favors the accumulation of immunosuppressive Tregs at the tumor site.
Recent studies suggested that cells express only PD-1
+ indeed retain Ag responsiveness in tumors, while only co-expression of PD-1
+ and Tim-3
+ identifies the most profoundly hypo-functional T-cells [
14‐
16], and the blockade of them rejuvenates tumor-infiltrating CD8
+ T-cells function in cancer patients. In present study, we compared the expression of PD-1
+ and Tim-3
+ on CD4
+/CD8
+ T-cells in blood circulation, tumor, and paraneoplastic tissues from patients with GC, and further tested the changes of IFN-γ production on T-cells after in vitro blockade of PD-1
+ and/or Tim-3
+ pathways. Our results showed that the frequencies of Tim-3
+, PD-1
+, and PD-1
+ Tim-3
+ cells among CD4
+/CD8
+ cells in circulation were significantly higher in GC patients than that in HDs. These data were consistent with previous reports of increased PD-1
+ and Tim-3
+ expression on T-cells in GC, which suggested that PD-1
+ and Tim-3
+ may be involved in immune evasion in GC patients [
30‐
33]. In addition, PD-1
+ and Tim-3
+ expression has been described in gastric patients with
Helicobacter pylori (
H. pylori) infection, which is a major cause of gastric cancer [
34,
35]. Studies showed that the removal of
H. pylori infection could theoretically decrease the number of cases by 89% [
36,
37]. Therefore, the interaction among
H. pylori infection, the expression of PD-1
+ and Tim-3
+, and GC needs to be further explored.
It was also observed that the percentages of PD-1
+, Tim-3
+, and PD-1
+ Tim-3
+ cells among CD4
+/CD8
+ T-cells were significantly increased in the tumor tissues compared to their counterparts in matched peripheral blood and paraneoplastic tissues. Meanwhile, the percentages of Tim-3
+, PD-1
+, and PD-1
+ Tim-3
+ cells among CD4
+ cells in paraneoplastic tissues were all significantly higher than those in peripheral blood. These results provided a solid foundation that TILs showed functional exhaustion in patients with GC, and supported the hypothesis that the tumor microenvironment played an important role in the up-regulation of inhibitory receptors [
16,
38,
39]. Furthermore, our data indicated that the inhibition of PD-1
+ and Tim-3
+ significantly enhanced tumor-infiltrating CD4
+/CD8
+ T-cells IFN-γ secretion in patients with GC compared with the control group. These results were concordant with previous reports of impaired T-cells during viral infections and tumor growth and suggested that co-expression of Tim-3
+ and PD-1
+ was a marker of tumor-induced T-cell dysfunction [
13,
38‐
40]. Previous researches have shown that the combination of Tim-3
+ blockade with PD-1
+ pathway blockade was remarkably more effective in colon carcinoma, acute myelogenous leukemia, and melanoma models than with blockade of either the Tim-3
+ or PD-1
+ pathway alone [
41,
42]. In this study, we also observed that combined PD-1
+ and Tim-3
+ inhibition had a synergistic effect on CD4
+ T-cells’ IFN-γ secretion, which was in an agreement with Smyth and Cunningham’s study [
43]. But combined PD-1
+ and Tim-3
+ inhibition did not have synergistic effects on IFN-γ induction of CD8
+ T-cells. This may be caused by the frequency of Tim-3
+ PD-1
+ T-cells occupying almost 90% of Tim-3
+ CD8
+ T-cells. In addition, although blockade of PD-1
+ or Tim-3
+ failed to improve the ability of nontumor-infiltrating CD4
+ T
-cells to produce IFN-γ, the combined PD-1
+ and Tim-3
+ inhibition had synergistic effects on IFN-γ induction of nontumor-infiltrating CD4
+ T-cells. These results could be explained by following points:
1.
The hierarchical co-regulation of multiple negative regulatory pathways on CD4
+ and CD8
+ T-cells [
44];
2.
The complex interactions between the inhibitory pathways during long-term in vitro conditions; and
3.
The blocking Tim-3/galectin-9 interactions complementary to PD-1
+ pathway inhibition [
45].