Tumor-infiltrating lymphocytes (TILs) were first described in 1863 by Robert Virchow and were found in proximity of the tumor.1 TILs are identified as a heterogenous cell group, include effector T cells, T regulatory cells, natural killer cells, macrophages, dendritic cells, myeloid-derived suppressor cells, and other immune cell types.1 The major populations of TILs that grow from tumors are CD8+ T cells and CD4+ T cells in vitro culture.2 The frequency of TILs was correlated with a better prognosis in patients with different types of tumors,3–5 including non–small lung cancer (NSCLC).6,7 The subtypes of TILs in NSCLC have been reported with different prognostic effect. Mori et al’s8 report indicates that the number of CD8+ TILs in NSCLC is not with a favorable prognosis; CD4+ T cells, not CD8+ T cells, in NSCLC cancer nests are associated with a favorable prognosis.9 Hiraoka et al’s10 study demonstrated that the infiltration of CD8+ and CD4+ TIL cells are a favorable prognostic factor in NSCLC. In 2010, another report shows that a high frequency of CD8+ TILs in NSCLC tissues is correlated with a favorable prognosis.11 In light of recent studies of immunotherapy, adoptive cell therapy (ACT) immunotherapy with autologous TILs with the memory phenotype yields drastic regression of malignant melanoma,12–15 whereas transferring terminal differentiation of TILs have poor antitumor immunity and short-term persistence.12,16 In other human advanced cancer (such as NSCLC), the efficacy of TIL therapy is uncertain. To further improve the antitumor effect and therapeutic potential of ACT with TILs in NSCLC, it is necessary to identify the composition and function of T cells in TILs.
According to the surface marker, the function and proliferation capacity, T cells can be categorized into naive T cells (Tn), effector T cells (Teff), and memory T cells (Tm); memory T cells include central memory T cells (Tcm), effector memory T cells (Tem), stem cell-like memory T cells (Tscm), and tissue-resident memory T cells (Trm).17,18 The distribution and function of Tscm cells in human lung cancer in the peripheral blood and lymph nodes have been studied.19 In this study, we investigated the characteristics of T cells in TILs from NSCLC. We found that CD3+ CD8+ CD45RA+ T cells outnumbered CD3+ CD4+ CD45RA+ T cells, whereas CD3+ CD4+ CD45RO+ T cells outnumbered CD3+ CD8+ CD45RO+ T cells. CD4+ Tem cells predominated in CD4+ TILs, the proportion of CD8+ Teff cells was higher than that of CD4+ Teff cells. About 12% of CD4+ Tscm and 10% of CD8+ Tscm were detected in TILs. To further analyze the function of T-cell subsets in TILs, we stimulated TILs from NSCLC patients with mitogens to examine cytokine production. Our data demonstrate that naive-phenotype T cells in TILs secret interferon-γ (IFN-γ) in abundance; tumor necrosis factor (TNF)-α-producing T cells were significantly increased, but fewer CD8+ Tscm cells produced TNF-α than CD4+ Tscm cells in TILs; there were more interleukin-17 (IL-17)-expressing CD4+ Tscm cells than other subtypes of CD4+ T cells in TILs. An accurate understanding of the subsets of T cells in TILs from NSCLC is critical for the prognosis and personalized medicine with TILs immunotherapy.
MATERIALS AND METHODS
A total of 18 NSCLC patients from the First Affiliated Hospital of Sun Yat-Sen University of Guangzhou, China, were enrolled in this study. These patients included 6 women and 12 men, their age ranging from 44 to 76 years. The final diagnosis of lung cancer, based on pathologic evidence (detected by histologic staining), included 6 cases of stage III, 3 cases of stage II, and 9 cases of stage I cancer (Table 1). Patients whose serology tested positive for HIV, HBV, and HCV were excluded from the study. None of the patients received cancer-related chemotherapy during the period of sample collection.
Written informed consent was obtained from all patients and healthy donors. This study was approved by the ethics committees of the Zhongshan School of Medicine, the Sun Yat-Sen University (Guangzhou, China), and the First Affiliated Hospital of Sun Yat-Sen University (Guangzhou, China).
Lung cancer tissues were maintained in cold Hanks’ buffer and brought to the laboratory within 2–4 hours after surgery; then blood, adipose tissues, and connective tissues of the tumor tissues were removed. The tumor tissues were cut into 1–2 mm3 pieces and plated into a 24-well plate containing complete X-VIVO 15 medium (cat. 04-418Q; Lonza, Walkersville, MD) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Invitrogen, Grand Island, NY), 100 U/mL penicillin (cat. 15071163), 100 mg/mL streptomycin (cat. 15071163), and 1000 UI/mL IL-2 (cat. 200-02; Peprotech, Rocky Hill, NJ). The medium was changed every 3 days, and the cells were extended after the medium turned yellow.
Flow Cytometry Analysis
The TILs from the lung cancer patients were stained for flow cytometry. The following panel of mouse anti-human mAbs, all purchased from BD Biosciences (San Jose, CA) or eBioscience (San Diego, CA), was used: anti-human CD3-APC.cy7 (BD, 557832, SK7), anti-human CD4-Percp.cy5.5 (BD, 560650, RPA-T4), anti-human CD45RO-FITC (eBioscience, 11-0457-42, UCHL1), anti-human CCR7-AF700 (BD, 561143, 150503), anti-human CD62L-PE-CF594 (BD, 562301, Dreg-56), anti-human CD95-PE.cy7 (BD, 561633, DX2), and anti-human CD122-PE (BD, 554525). The cell data were acquired using a BD LSRFortessa analytical flow cytometer. Unstained and single fluorochrome-stained cells were used as controls to provide accurate compensation and data analysis. Cells were counted per sample and the data were analyzed with FlowJo (Version 10).
The TILs were incubated in 24-well plates at 2×106 cells per well in RP10 media (RPMI, 10% heat-inactivated FBS) alone or with PMA (20 ng/mL) plus ionomycin (1 μg/mL) for 4–6 hours at 37°C in the presence of BFA (10 μg/mL). The cells were harvested, washed with PBS, stained for the surface phenotypic markers, and fixed at RT with 4% PFA (Sigma). The cells were then permeabilized (0.01% saponin), and the intracellular cytokines were stained using anti-human IFN-γ-V450 (BD, 560371, B27), anti-human IL-17A-APC (BD, 560486, N49-653), and anti-human TNF-α-APC (eBioscience, 17-7349-82,MAB11). All samples were analyzed using a BD LSRFortessa instrument. The data were analyzed using the FlowJo software. Phorbol 12-myristate13-acetate (PMA) (cat. 16561-29-8), ionomycin (cat. 10634), brefeldin A (BFA) (cat. B7651), bovine serum albumin, and NaN3 were all purchased from Sigma-Aldrich (St Louis, MO).
GraphPad Prism software version 5 was used for the statistical analysis. The Mann-Whitney test (2-tailed) and the nonpaired Student t test were performed to identify significant differences. A value of P≤0.05 was considered statistically significant.
The Frequency of CD45RA+ and CD45RO+ T Cells in TILs of Human Lung Cancer
Eighteen lung cancer patients recruited to this study had been diagnosed with non–small cell lung cancer (NSCLC) and were HIV and HBV negative and free of other cancers. T lymphocytes infiltrating tumor deposits were cultured from lung cancer tissues.
CD45RO is a well-known surface marker to define memory T cells; CD45RA expression is designated in naive and terminal effector T cells.20 To assess the frequency of CD45RA+ and CD45RO+ T cells in TILs from NSCLC, we analyzed the CD3+ CD4+ CD45RA+/CD45RO− T cells, CD3+ CD4+ CD45RO+ T cells, CD3+ CD8+ CD45RA+/CD45RO− T cells, and CD3+ CD8+ CD45RO+ T cells in TILs by flow cytometry. The marker expression of naive and memory T cells was determined by representative flow cytometry analysis of CD45RO expression in CD3+ CD4+ and CD3+ CD8+ T cells, respectively (Fig. 1A, Supplementary Fig. a, Supplemental Digital Content 1, http://links.lww.com/JIT/A454). The analysis of CD3+ CD4+ T cells and CD3+ CD8+ T cells indicated that TILs contained similar frequencies (P=0.9212) (Fig. 1B). The proportion and absolute count of the CD3+ CD4+ CD45RO+ T cells were higher than that of CD3+ CD4+ CD45RA+ T cells (P<0.0001) (Fig. 1C); CD4+ CD45RO+ T cells predominated in CD4+ TILs. In contrast, CD45RA+ T cells and CD45RO+ T cells possessed a comparable proportion of CD8+ TILs (Figs. 1C, D) (P=0.9107), whereas the absolute number of CD8+ CD45O+ T cells was lower than that of CD8+ CD45RA+ T cells, but there was no significant difference (P=0.1256) (Fig. 1C). In CD45RA+ TILs from NSCLC, CD8+ CD45RA+ T cells outnumbered CD4+ CD45RA+ T cells (P<0.0001), whereas in CD45RO+ TILs, CD4+ CD45RO+ T cells outnumbered CD8+ CD45RA+ T cells (P=0.0027) (Fig. 1D). Our results indicate that the frequency of CD45RA+ and CD45RO+ T cells is different between CD4+ and CD8+ TILs from human lung cancer.
The Frequency of Subsets of TILs From Human Lung Cancer
In human beings, there are 2 distinct subsets of memory T cells based on the expression of lymph node homing receptor CCR7. CD45RO+ CCR7+ were designated Tcm cells; Tem cells were CD45RO+ CCR7−; CD45RA+ T cells are classified into 2 subsets: CCR7+ Tn cells and CCR7− Teff cells.20,21 To further analyze the subsets of TILs, we determined the Tn, Tcm, Tem, and Teff cell populations of TILs from human lung cancer by flow cytometry according to the established surface markers (Fig. 2A). CD4+ Tem cells in TILs were present at a high frequency (56.08%), followed by CD4+ Tcm (21.85%) and CD4+ Teff (14.67%) cell subsets; CD4+ Tn cells (7.41%) existed at low frequencies (Fig. 2B) (P=0.0045). However, the composition of CD8+TILs differed from that of CD4+ TILs, CD8+ TILs contained Tem cells (37.65%) and Teff cells (33.99%) in similar proportion, which were greater than those of Tn cells (16.62%) (P=0.0036), and there was a small fraction of Tcm cells (11.75%) in CD8+ TILs (Fig. 2B). CD4+ Tem cells were at a higher proportion than CD8+ Tem cells in TILs (P=0.0049), whereas the frequency of CD8+ Teff was higher than that of CD4+ Teff cells (P=0.0066) (Fig. 2C). These results reveal that the frequency of T-cell subsets differ between CD4+ and CD8+ TILs from NSCLC patients.
Identification of Tscm in TILs From Human Lung Cancer
Tscm cells have been recently identified as a new population of memory T cells; they possess the capacity to self-renew and differentiate into Tcm, Tem, and Teff cells, and Tscm cells exhibit strong antitumor function with adopting transfer different memory T cells subsets in mouse melanoma model.22 Because of the increased proliferative capacity and antitumor activity, Tscm cells can be estimated to be an optimal cell type for ACT to improve the long-term persistence of antitumor. According to the surface marker of Tscm cells, CD3+ CD4+/CD8+ CD45RA+ CD62L+ CCR7+ CD27+ CD28+ CD127+ CD95+ CD122+ Tscm cells have been identified in human PBMCs,22 and we have investigated the distribution and function of Tscm cells in the blood and lymph nodes from NSCLC patients,19 but the frequency and function of Tscm cells in TILs from NSCLC remains unknown. We found that CD27, CD28, and CD127 were negative in TIL cells from human lung cancer patients. In the representative flow cytometric analysis on CD3+CD4+/CD8−CD45RA+CD62L+CCR7+CD95+CD122+ marker for CD4+ Tscm cells or CD3+CD4−/CD8+CD45RA+CD62L+CCR7+CD95+CD122+ marker for CD8+ Tscm cells (Fig. 3A), because the color combination is not allowed in the flow cytometry equipment, we compared the frequency of CD3+CD4−/CD8+T cells (47.3%) and CD3+CD8+/CD4− T cells (41.9%) in TILs from the same patient (Supplementary Fig. b, Supplemental Digital Content 1, http://links.lww.com/JIT/A454). The frequency of CD4+ Tscm cells was 12.08% in TILs of NSCLC, higher than in blood of NSCLC (NSCLC-PBMC) (2.59%), and lower than in lymph nodes of NSCLC (NSCLC-Ly) (29.91%); the fraction of CD8+ Tscm cells was 9.87% in TILs, which was between that in NSCLC-PBMC (1.5%) and NSCLC-Ly (18.99%) (Fig. 3B).19 The proportion of CD4+ Tscm cells and CD8+ Tscm cells was similar (P=0.6195) (Fig. 3B). Our results demonstrate that Tscm cells can be detected in TILs from NSCLC patients and there is no appreciable difference in the proportion of CD4+ Tscm and CD8+ Tscm in TILs.
Tn Cells in TILs of Human NSCLC Patients Have a Strong Capacity to Produce IFN-γ
IFN-γ is dispensable for mediating antitumor response by recruiting antigen-specific CD8+ cytotoxic T lymphocytes into tumor or by inducing MHC class I expression.23 IFN-γ-secreting helper T cells (Th1) play a critical role in maintaining antitumor memory. Teff cells, memory T cells, and Tscm cells produce IFN-γ after stimulating with specific antigen or nonspecific agents, whereas Tn cells have a limited capacity to produce IFN-γ. We have observed IFN-γ production in Tn cells from PBMC and lymph nodes from human lung cancer: the proportion of IFN-γ-secreting Tn cells was no more than 4% both in CD4+ Tn and CD8+ Tn cells, and the composition was not significantly different from that of healthy donors.19 To examine whether the capacity of IFN-γ-secreting Tn cells was different in TILs from human lung cancer, we simulated TILs with PMA+ionomycin for 4 hours and investigated that >50% of Tn cells produce IFN-γ; IFN-γ-secreting CD8+ Tn cells were remarkably similar between CD8+ Tcm cells (64.99%), CD8+ Tem cells (72.06%), CD8+ Teff cells (53.94%), and CD8+ Tscm cells (53.94%) in TILs (P>0.05); IFN-γ-producing CD4+ Tn cells (54.31%) were significantly lower than CD4+ Tscm cells (P=0.016) (Figs. 4A, B); Tscm cells (77.61%) represented a preponderance of IFN-γ producers, followed by Tem cells (73.49%), Teff cells (72.22%), and Tcm cells(67%) in CD4+ TILs (Fig. 4B), but no difference was observed in the fraction of IFN-γ expressing among the CD4+ Tscm cells, CD4+ Tcm cells, CD4+ Tem cells, and CD4+ Teff cells (P>0.05) (Figs. 4A, B). There were fewer IFN-γ-secreting Tscm cells in lymph nodes than in TILs from NSCLC, a high frequency CD4+ Tscm IFN-γ producer in TILs than in blood, a similar frequency of IFN-γ-expressing CD8+ Tscm cells was observed between TILs and blood from NSCLC.19 These results indicate that naive-phenotype T cells in TILs have a strong ability to produce IFN-γ.
TNF-α Production is Significantly Increased in Subsets of TILs of NSCLC Patients
TNF-α) is a proinflammatory cytokine, which mediates anticancer adaptive immune response.24 TNF-α production in various subsets of TILs from human lung cancer needs to be further studied. TNF-α producers in TILs from human lung cancer represented approximately 60% of the CD4+ Tn cells (59.1%), which was significantly lower than the levels observed in CD4+ Tcm cells (77.5%) (P=0.0287), CD4+ Tem cells (88.44%) (P=0.0006), and CD4+ Teff cells (72.66%) (P=0.0272), and we observed the same pattern for CD8+ TILs. The proportion of TNF-α-expressing CD8+ Tscm cells (46.73%) was lower than that of CD8+ Tem cells (76.31%) (P=0.0087) and CD8+ Teff cells (72.55%) (P=0.0303) (Figs. 5A, B), and the fraction of TNF-α-producing CD8+ Tn cells (49.87%) was also lower than that of CD8+ Tem cells (P=0.0016) and CD8+ Teff cells (P=0.0142). In contrast, TNF-α expressor among the subsets of CD4+ and CD8+ T cells in TILs were found to exceed those of the CD4+ and CD8+ T-cell subgroup (no more than 20%) in blood and lymph node from NSCLC.19 In TILs of the NSCLC patients, the proportion of TNF-α-expressing CD4+ Tscm cells was higher than that of the CD8+ Tscm cells (Fig. 5B) (P=0.0063). In contrast, the frequency of TNF-α-expressing Tscm cells in TILs was higher than that in the blood or lymph nodes of NSCLC. Together, these results suggest that TNF-α production is significantly increased in TILs of NSCLC patients.
IL-17-expressing CD4+ Tscm Cells are Remarkably Increased in TILs of NSCLC Patients
IL-17 (IL-17 cells, also known as IL-17A) is a member of the IL-17 family of cytokines; IL-17 is primarily produced by CD4+ helper T cells (Th17 cells). There is no controversy about the contribution of Th17 to inflammation and autoimmunity, but contradictions remain regarding the role of IL-17 in tumor immunity. Researches showed that Th17 cells and IL-17-producing CD8+T cells possess stem cell-like traits in mice and humans.25,26 We previously observed that a low fraction of IL-17-producing memory CD4+ T cells (CD4+ Tcm and CD4+ Tem cells) were present in PBMCs and lymph nodes from human lung cancer; the expression of IL-17 was not detectable in CD8+ T cells and other subsets of CD4+ T cells.19 After stimulation with PMA+ionomycin, we observed a high proportion of IL-17-expressing CD4+ Tscm cells (14.8%) in TILs from human lung cancer, which was greater than that of IL-17 production in CD4+ Tn cells (5.31%) (P=0.0391), CD4+ Tcm cells (4.13%) (P=0.0224), CD4+ Tem cells (4.66%) (P=0.0332), and CD4+ Teff cells (3.42%) (P=0.0134). The frequency of IL-17-secreting CD4+ TILs exceeded that of CD4+ T cells in PBMCs and lymph nodes from human lung cancer. In CD8+ TILs, CD8+ Tcm cells (1.56%) and CD8+ Tem cells (2.01%) existed as moderate IL-17 producers, minimal IL-17 was produced by CD8+ Tn cells (0.91%) and CD8+ Teff cells (0.99%), whereas we did not detect IL-17 production in CD8+ Tscm cells (Fig. 6B). These results indicate that CD4+ Tscm cells in TILs from NSCLC have a strong capacity to produce IL-17 in TILs of NSCLC.
In this study, we analyzed the classification and function of the CD4+ and CD8+ T cells in TILs of human NSCLC patients. CD4+ CD45RO+ T cells are the primary cell type in CD4+ TILs: half CD45RA+ T-cell and CD45RO+ T-cell population in CD8+ TILs. In the CD45RA+ T-cell population, we found that the frequency of the CD4+ Tn cells (7.41%) was lower than that of CD8+ Tn (16.62%); the proportion of CD4+ Teff cells (14.67%) was also smaller than that of CD8+ Teff (33.99%); Tscm cells shared both the naive and the memory T-cell phenotype; there was a similar fraction of CD4+ Tscm cells (12.08%) and CD8+ Tscm cells (9.87%) in TILs. In the functional analysis, we found that CD4+ Tn cells and CD8+ Tn cells in TILs can produce a large amount of IFN-γ (>50%), whereas in other lymphoid sites, such as blood or lymph node, IFN-γ-producing Tn cells were under 10% from NSCLC patients or healthy donors. The same phenomenon was observed in TNF-α-expressing CD8+ Tn cells in TILs. As it is known, Tn cells have a very limited ability to produce cytokines, but Tn cells in TILs have a distinct capacity to secrete functional cytokines, we speculated that Tn cells in TILs are not “real” Tn cells. Therefore, further studies are needed.
As the subsets of memory T cells, Tscm cells possess the properties of self-renewal and multipotency; Tscm cells have the ability to regenerate and differentiate into the specialized lymphocytes against tumor; the existence of Tscm cells in TILs will enhance the curative effect based on the TIL immunotherapy. The frequency of Tscm cells in TILs was in between that of blood and lymph nodes of NSCLC. The relationship between Tscm and Tn cells in TILs need to be determined. Whether the CD45RA+ T cells contain other T-cell populations that are different from Tscm cells remains unknown. We observed that Tscm cells in TILs did not express the memory marker CD27, CD28, CD127, which is different from the Tscm cells in the blood and lymph nodes.21 CD27 (TNFRSF7) is one of the TNFRSF members that is involved in cellular activation.27 CD27 was identified to be expressed in peripheral blood CD4 and CD8 T cells after being activated.28 A research demonstrated that CD27 could be lost on some memory phenotype T cells if activated T cells were exposed to antigen stimulation for a long term.29 TILs infiltration in tumor tissues with potential chronic repeated stimulation by the tumor antigen might lead to the downregulation of CD27 expression; another possibility is that under the culture condition in vitro, there may be a loss of CD27 expression. The first costimulatory molecule is CD28, which is required for T-cell activation. CD27 and CD28 are expressed on early memory T cells, Tn cells, Tscm cells, and Tcm cells, whereas effector memory T cells, effector T cells, and exhausted T cells show low-level expression or loss.22,30,31 However, in TILs, we did not detect CD27 or CD28 expression on the early memory Tscm cells.
In an acute viral infection of mouse model, CD127 (IL-7Rα, IL-7 receptor α-chain) was a marker that distinguished effector T cells from memory T cells and was directly associated with CD8+ effector T-cell survival, and CD127highCD8+ T cells were identified as memory T-cell precursors after infection.32 Another research showed that CD127 together with CD45RO−, CCR7+, CD45RA+, CD62L+, CD27+, CD28+ were identified as characteristic of Tn cells, and CD127 was one of the Tscm cell markers.22 Tscm cells from TILs are CD127 negative, which is different from the Tscm cells from blood and lymph nodes.
About the functional attributes, IFN-γ, TNF-α-producing Tscm cells in TILs are much higher than in blood and lymph nodes from NSCLC, whereas the frequency of IFN-γ-producing CD8+ Tscm cells in TILs is similar to that of in blood from NSCLC. The most striking functional characteristic of Tscm cells in TILs is IL-17 production. Associated stemness genes (such as Tcf7 and β-catenin) are highly expressed on Tscm cells, Th17 cells, and IL-17-producing CD8+ T cells, displaying some stem cell characteristics. Th17 cell could give rise to IFN-γ-secreting Th1 cells.25,26 Recent studies show that ACT using Th17 cells can enhance the long-term antitumor response.26 In TILs, we found that all subsets of CD4+ T cell could produce IL-17, but the IL-17 level was remarkably increased in CD4+ Tscm cells. In contrast to other lymphoid tissue from NSCLC, CD4+ TILs secrete large amounts of cytokine. IL-17-producing CD8+ T cells in TILs were CD8+ Tcm cells and CD8+ Tem cells, whereas in CD8+ Tscm, no IL-17 production was detected. Together, our findings demonstrate that the frequencies of subpopulation in TILs from NSCLC are different, and each group of TILs has distinct functional properties. Better understanding of the immunobiology of TILs will increase our ability to manipulate tumor-reactive TILs for more effective and therapeutic treatment of cancer patients.
The authors thank other members of the laboratory for their assistance.
CONFLICTS OF INTEREST/ FINANCIAL DISCLOSURES
This study was in part supported by Guangzhou Science Foundation of China (42020075), the National Basic Research Program of China (973 Program) (33000-41080962), and Guangdong Innovative Research Team Program (201001Y01046872443). R.F.W was in part supported by grants (CA101795 and DA030338) from NCI, NIDA, and NIH.
All authors have declared there are no financial conflicts of interest with regard to this work.
1. Mantovani A, Allavena P, Sica A, et al. Cancer-related inflammation. Nature. 2008;454:436–444.
2. Rosenberg SA, Restifo NP. Adoptive cell transfer as personalized immunotherapy for human cancer. Science. 2015;348:62–68.
3. Loi S, Michiels S, Salgado R, et al. Tumor infiltrating lymphocytes are prognostic in triple negative breast cancer and predictive for trastuzumab benefit in early breast cancer: results from the FinHER trial. Ann Oncol. 2014;25:1544–1550.
4. Azimi F, Scolyer RA, Rumcheva P, et al. Tumor-infiltrating lymphocyte grade is an independent predictor of sentinel lymph node status and survival in patients with cutaneous melanoma. J Clin Oncol. 2012;30:2678–2683.
5. Lee N, Zakka LR, Mihm MC Jr, et al. Tumour-infiltrating lymphocytes in melanoma prognosis and cancer immunotherapy. Pathology. 2016;48:177–187.
6. Horne ZD, Jack R, Gray ZT, et al. Increased levels of tumor-infiltrating lymphocytes
are associated with improved recurrence-free survival in stage 1A non-small-cell lung cancer
. J Surg Res. 2016;171:1–5.
7. Velcheti V, Schalper KA, Carvajal DE, et al. Programmed death ligand-1 expression in non-small cell lung cancer
. Lab Invest. 2014;94:107–116.
8. Mori M, Ohtani H, Naito Y, et al. Infiltration of CD8+T cells in non-small cell lung cancer
is associated with dedifferentiation of cancer cells, but not with prognosis. Tohoku J Exp Med. 2000;191:113–118.
9. Wakabayashi O, Yamazaki K, Oizumi S, et al. CD4+ T cells in cancer stroma, not CD8+ T cells in cancer cell nests, are associated with favorable prognosis in human non-small cell lung cancers. Cancer Sci. 2003;94:1003–1009.
10. Hiraoka K, Miyamoto M, Cho Y, et al. Concurrent infiltration by CD8+ T cells and CD4+ T cells is a favourable prognostic factor in non-small-cell lung carcinoma. Br J Cancer. 2006;94:275–280.
11. Zhuang X, Xia X, Wang C, et al. A high number of CD8+ T cells infiltrated in NSCLC tissues is associated with a favorable prognosis. Appl Immunohistochem Mol Morphol. 2010;18:24–28.
12. Crompton JG, Sukumar M, Roychoudhuri R, et al. Akt inhibition enhances expansion of potent tumor-specific lymphocytes with memory cell characteristics. Cancer Res. 2015;75:296–305.
13. Chodon T, Comin-Anduix B, Chmielowski B, et al. Adoptive transfer of MART-1 T-cell receptor transgenic lymphocytes and dendritic cell vaccination in patients with metastatic melanoma. Clin Cancer Res. 2014;20:2457–2465.
14. Rosenberg SA, Yang JC, Sherry RM, et al. Durable complete responses in heavily pretreated patients with metastatic melanoma using T-cell transfer immunotherapy. Clin Cancer Res. 2011;17:4550–4557.
15. Wu R, Forget MA, Chacon J, et al. Adoptive T-cell therapy using autologous tumor-infiltrating lymphocytes
for metastatic melanoma: current status and future outlook. Cancer J. 2012;18:160–175.
16. Deplancke B, Romero P, Rufer N, et al. Exhaustion of tumor-specific CD8+
T cells in metastases from melanoma patients. J Clin Invest. 2011;121:2350–2360.
17. Kaech SM, Cui W. Transcriptional control of effector and memory CD8+T cell
differentiation. Nat Rev Immunol. 2012;12:749–761.
18. Gattinoni L, Klebanoff CA, Restifo NP. Paths to stemness: building the ultimate antitumour T cell
. Nat Rev Cancer. 2012;12:671–684.
19. Hai H, Yong G, Si Yuan SH, et al. The distribution of human stem cell-like memory T cell
in lung cancer
. J Immunother. 2016;39:233–240.
20. Sallusto F, Lenig D, Forster R, et al. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature. 1999;401:708–712.
21. Masopust D, Vezys V, Marzo AL, et al. Preferential localization of effector memory cells in nonlymphoid tissue. Science. 2001;291:2413–2417.
22. Gattinoni L, Lugli E, Ji Y, et al. A human memory T cell
subset with stem cell-like properties. Nat Med. 2011;17:1290–1297.
23. Wilcox RA, Flies DB, Wang H, et al. Impaired infiltration of tumor-specific cytolytic T cells in the absence of interferon-gamma despite their normal maturation in lymphoid organs during CD137 monoclonal antibody therapy. Cancer Res. 2002;62:4413–4418.
24. Guiducci C, Vicari AP, Sangaletti S, et al. Redirecting in vivo elicited tumor infiltrating macrophages and dendritic cells towards tumor rejection. Cancer Res. 2005;65:3437–3446.
25. Muranski P, Borman ZA, Kerkar SP, et al. Th17 cells are long lived and retain a stem cell-like molecular signature. Immunity. 2011;35:972–985.
26. Kryczek I, Zhao E, Liu Y, et al. Human TH17 cells are long-lived effector memory cells. Sci Transl Med. 2011;3:104ra100.
27. Croft M, Benedict CA, Ware CF. Clinical targeting of the TNF and TNFR superfamilies. Nat Rev Drug Discov. 2013;12:147–168.
28. Bigler RD, Bushkin Y, Chiorazzi N. S152 (CD27). A modulating disulfide-linked T cell
activation antigen. J Immunol. 1988;141:21–28.
29. Hintzen RQ, de Jong R, Lens SM, et al. Regulation of CD27 expression on subsets of mature T-lymphocytes. J Immunol. 1993;151:2426–2435.
30. Berger C, Jensen MC, Lansdorp PM, et al. Adoptive transfer of effector CD8+ T cells derived from central memory cells establishes persistent T cell
memory in primates. J Clin Invest. 2008;118:294–305.
31. Hinrichs CS, Borman ZA, Cassard L, et al. Adoptively transferred effector cells derived from naive rather than central memory CD8+ T cells mediate superior antitumor immunity. Proc Natl Acad Sci USA. 2009;106:17469–17474.
32. Kaech SM, Tan JT, Wherry EJ, et al. Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. Nat Immunol. 2003;4:1191–1198.