Lung cancer remains the leading cause of cancer-related death worldwide [1
]. Although small improvements have been made in the treatment of non-small cell lung cancer (NSCLC) over past decades, the efficacy of the new immunotherapy methods targeting immune checkpoints has been recently demonstrated in about 40% of patients [3
]. More accurate biomarkers for individual therapy are needed.
Cancer stem cells (CSCs) are small numbers of cells that exist in the tumor microenvironment (TME) and hold stemness properties that sustain cancer progression, such as enhanced capacities for self-renewal cloning, growing, metastasizing, homing, and reproliferating [5
]. Lung cancer TME is composed of a complex group of noncancer cells, such as stromal cells, tumor-associated macrophages, tumor-infiltrating lymphocytes, regulatory T cells, myeloid-derived suppressor cells, dendritic cells, NK cells, and natural killer T cells along with cancer cells: CSCs and mature tumor cells (MTCs). There are different accepted definitions for MTCs and CSCs regarding the surface markers they express. In the current state of knowledge, lung MTCs are described as CD45-/EpCAM+/CXCR4+ [6
]. Lung CSCs are defined as CD45-/EpCAM+/CXCR4+/CD44+/CD133+/CD90+ [8
]. It is considered that CSCs are responsible for recruiting noncancer cells to TME [12
]. However, the interplay between CSCs and noncancer cells in TME, as well as interactions of CSCs with the immune system in the systemic circulation, is not fully understood. While there have been enormous advances in the understanding of the immunoediting process in the TME and its prognostic significance, less is known about changes in the regulatory mechanisms at the level of the LNs [13
]. LNs are common sites of metastasis. Cancer cells that have metastasized to LNs may express immunosuppressive molecules and escape immune detection. Similar to primary TME, MTCs in metastatic LNs shape their interactions with the host immune system by controlling the infiltration and reactivity of immune cells [14
]. At present, little is known about the frequency of immunomodulatory molecules on MTCs and CSCs in metastatic LNs in NSCLC patients.
As tumors develop, they acquire mechanisms to avoid attenuation by the immune system [15
]. T cells play a key role in anticancer defense, but their population is modulated during the development of the disease [16
]. The numerous suppressory and regulatory mechanisms inhibit the recognition of lung cancer antigens and are capable of blocking lymphocyte activation [17
]. The strong suppressory pathway is the programmed death-1/programmed death ligand-1 (PD-1/PD-L1) pathway. Understanding these mechanisms can lead to the development of new strategies that provoke the immune system to recognize lung cancer as foreign. The success of checkpoint inhibitors creates interest in evaluating other antibodies that can be used to modulate T cell responses and/or enhance the ability of the immune system to destroy the tumor cells.
In our previous study, we confirmed the presence of PD-L1+ CSCs in metastatic LNs in NSCLC patients that may suggest their immunosuppressive properties [18
]. This study aimed to evaluate the presence of immunosuppressing molecules: PD-L1, CD47, CD73, Fas (CD95), and FasL (Fas Ligand) on CSCs and MTCs in LNs aspirates, and to evaluate the presence of PD-1 (programmed death receptor 1), Fas, and LAG3 (Lymphocyte-activation gene 3) on CD4 T cells and CD8 T cells in systemic circulation in NSCLC patients (Figure 1
); based on the examination of the endobronchial ultrasound-guided transbronchial needle aspiration (EBUS/TBNA) technique and the peripheral blood (PB) from the same patient.
We proved in our current study that the cellular composition of LNs could be measured during standard-of-care bronchoscopic assessment using flow cytometry. In contrast to standard immunohistochemistry, flow cytometry enables the simultaneous comprehensive analysis of many immunomodulatory markers in a single-cell population. The usefulness of this technique in the assessment of LNs in NSCLC patients has been confirmed by other authors [20
]. For the first time, we detected CSCs and also MTCs, with the expression of molecules capable of modulation of the immune response, in LNs and PB of lung cancer patients. The interaction between CSCs and the immune system is not well understood and is currently of much interest. In our previous study, we confirmed the presence of PD-L1+ CSCs in LNs aspirates in NSCLC patients suggesting their immunogenic potential [18
]. It encouraged us to investigate the presence of other immunomodulatory molecules on lung CSCs. The antibody panel we designed allows defining both CSC and MTC populations. We found that both CSCs and MTCs express all investigated immunomodulatory molecules: PD-L1, CD47, FasL, Fas, and CD73. The presence of these molecules was described on lung cancer cells and is adopted as a biomarker to immunotherapy [22
]. To date, in the case of lung CSCs, only the presence of PD-L1 and CD47 has been presented [18
We found that among all investigated immunomodulatory molecules, PD-L1 and CD47 have the highest expression (GMF) on CSCs and MTCs. The importance of PD-L1 for cancer immunology and treatment has become widely known. No defined clinical data are available in regard to CD47, but this molecule has generated significant interest to date [26
]. In general, PD-L1 is a critical “do not find me” signal to the adaptive immune system [25
], whereas CD47 is a critical “do not eat me” signal to the innate immune system, as well as a regulator of the adaptive immune response [27
]. CD47 ablation stimulates macrophage phagocytosis and polarization and synergizes with PD-1 blockade [26
]. Furthermore, anti-CD47 antibody or CD47 blockade treatments have been demonstrated to reduce tumor burden and increase patient survival in various tumor xenograft models [28
Interestingly, we observed that GMF of both CD47 and PD-L1 was higher on CSCs than on MTCs. These results are in concordance with the study of Liu et al. [25
], who reported the higher CD47 expression on CSCs than on tumor cells in lung cancer cell lines, using flow cytometry. One explanation might be that MTCs lose expression of some molecules during differentiation. The knockdown of CD47 suppressed certain stem-like properties of cancer cells, such as self-renewal and chemoresistance, suggesting that targeting CD47 could not only activate the phagocytosis of macrophages but also could be used to enhance treatment against CSCs [29
]. Hence, CD47 expression may be another mechanism used by lung cancer cells, especially lung CSCs, to escape phagocytosis. These results may indicate CD47 as a therapeutic target in lung cancer, as well as a therapeutic target in lung CSCs.
We noticed the presence of CD73 on the CSCs and MTCs population. The expression of CD73 in the tumor TME has been described in various types of cancer and is at least partly driven by hypoxia and activation of hypoxia-inducible (HIF) transcription factors [30
]. Hypoxia drives expression of the well-defined transcription factor HIF1α, which promotes the expression of ectoenzymes CD39 and CD73 on tumor cells, stromal cells, and tumor-infiltrating immunosuppressive cell subsets, such as regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs) [31
]. CD39 catalyzes the conversion of ATP and ADP into AMP, while CD73 catalyzes the irreversible conversion of AMP into adenosine [31
]. CD73-derived adenosine accumulates in the TME and exerts multiple immunosuppressive actions to dampen antitumor immunity, leading to worse clinical outcomes [30
]. Furthermore, CD73 has been shown to be biomarkers of patient outcomes in several tumor types, including NSCLC [23
]. It has been described that CD73 promotes the expression of stemness and epithelial-mesenchymal transition (EMT), implying a regulation of CSCs function in ovarian cancer [32
]. Anti-CD73 antibodies were shown to reduce tumor growth and metastasis through the activation of NK and T cell responses [31
]. Recently, CD73 has been an extensively investigated target in anticancer therapy and shows synergy with anti-PD-1/PD-L1 agents [33
The Fas/FasL system is involved in programmed cell death. Fas bearing cells are susceptible to apoptosis induced by connection with Fas ligand (FasL). It was observed that tumor cells that express FasL induce apoptosis of lymphocytes expressing Fas. On the other hand, cancer cells are resistant to apoptosis, and the Fas/FasL system is impaired. In our study, we have found that lung cancer cells express both Fas and FasL, which is in concordance with a study by Li et al. [34
]. The vast majority of reports demonstrate that the expression of both Fas and especially of FasL acts as a negative prognostic marker for many cancers [24
]. In our group, the reduced Fas expression on MTCs was found more frequently in the advanced stage than in the earlier stages. It also appeared to be significantly associated with higher nodal status.
What is more, we found that a FasL/Fas ratio was significantly higher in patients with confirmed metastases in LNs than in patients without metastatic disease. These associations suggest that defects in the apoptotic pathway represent a significant element in the progression of the NSCLCs. Altogether, these data support a role for the loss of Fas-mediated apoptosis during tumorigenesis and tumor progression.
The best to our knowledge, we report the presence of Fas and FasL on lung CSCs for the first time. The concept that Fas can be a tumor promoter has now gained wide acceptance, supported by several reports describing the marked activities of Fas in tumor growth and spread [24
]. Moreover, and related to this, Fas is capable of inducing the EMT process in gastrointestinal and breast cancer [35
]. In both studies, stimulation of Fas on cancer cells induced a conversion from non-CSCs to CSCs, in consequence, increasing the frequency of CSCs. Unfortunately, the molecular mechanisms underlying the switch between these different signaling pathways remain enigmatic.
FasL has a range of tumor-promoting activities, some of which are indirect, such as the suppression of the immune response in the cancer microenvironment, by killing Fas positive immune cells [19
]. We examined the presence of Fas on CD3 T cells in LNs and the presence of Fas on CD4 and CD8 T cells in PB. We found that Fas+ CD3 T cells were significantly correlated with FasL+ MTCs in LNs aspirates. There was a higher frequency of Fas+ CD8 T cells in PB in NSCLC patients with confirmed metastases than in patients without metastases. An elevated proportion of PB lymphocytes with Fas expression was previously reported in patients with lung cancer and COPD [37
]. Other authors have reported the higher percentage of Fas+ CD8 T in malignant pleural effusion (what may represent the tumor milieu), but not PB in lung cancer patients [39
]. It should be noted that the proportion of Fas+ lymphocytes is usually correlated with the intensity of tobacco exposure that leads to chronic inflammation [37
]. Here, we did not observe any significant correlation between the frequency of Fas+ lymphocytes and pack-years smoked.
In our previous study, we found some correlation between suppressory immune cells in LNs and PD-L1+ CSC [40
]. Here, we did not observe any significant dependencies between PD-L1+ CSCs, CD47+ CSCs, CD73+ CSCs population, and immunophenotype of lymphocyte in PB. In all, it seems that all examined immunomodulatory molecules may be involved in different pathways leading to tumor escape from immune surveillance, and each of them requires further investigation.
Finally, we found that among 7 samples classified as nonmetastatic in the histopathological examination, MTCs were found in 4 samples and CSCs in 3 samples in multiparameter flow cytometry assay; among these four patients, three were at the IIB stage of NSCLC and one at the IB stage of NSCLC. Immunohistochemistry remains the ‘gold standard’ in assessing LNs involvement [41
]. Research by other authors also showed the feasibility of EBUS/TBNA samples of LNs for flow cytometric analysis of MTCs [42
]. Moreover, a retrospective study performed by Gwozdz et al. demonstrated that the presence of occult micrometastases in the mediastinal LNs was associated with reduced survival in I and II stage NSCLC patients due to tumor recurrence [43
]. Their and our study demonstrated the usefulness of the EpCAM marker for the detection of early LNs cancer invasion.
We are aware that the investigated group is low. The sample size did not allow us to perform a reliable comparison between the histological subtypes of NSCLC and between patients with confirmed genetic alterations mutations. Another weakness of our study was the lack of follow-up and comparison of our findings to the course of the disease. Thus our results indicate and support the direction for further investigations.
4. Materials and Methods
We obtained 2 ml of peripheral blood (PB) and placed it in tubes containing K2EDTA and processed for flow cytometry.
LNs group 4, 7, 10, and 11 aspirates were obtained during routine EBUS/TBNA procedure of lung cancer diagnosis. After diagnostic aspiration, the additional sample was taken for flow cytometry analysis. About 1 ml of LNs aspirate was diluted in 0.9% NaCl, collected in tubes containing K2EDTA, and processed for flow cytometry.
To identify MTCs flow cytometry and staining using monoclonal antibodies targeting the cell-surface expression of CD45 (V450), EpCAM (FITC) and CXCR4 (APC) were used (BD, USA). MTCs were defined as CD45-/EpCAM+/CXCR4+. CSCs population was determined by a panel of monoclonal antibodies: EpCAM (FITC), CD133 (PE), CD90 (PE-Cy7), CXCR4 (APC), CD44 (APC-H7), and CD45 (V500). CSCs were defined as: CD45-/CXCR4+/EpCAM+/CD133+/CD44+/CD90+.
Additionally, anti-PD-L1 (PerCP-Cy5.5), anti-CD47 (PerCP-Cy5.5), anti-CD73 (PerCP-Cy5.5), anti Fas (PerCP-Cy5.5), and FasL (V450) antibodies were applied to assess the presence of these molecules on MTC and CSCs in LNs aspirates.
The proportion of CD4+ or CD8+ subpopulations in PB and LNs aspirates were determined by a panel of monoclonal antibodies: anti CD45 (V500), anti CD3 (APC-H7), anti CD8 (V450), and anti CD4 (PE-Cy7).
Additionally, anti-PD-1 (FITC), anti-Fas (PerCP-Cy5.5), and anti-Lag3 (PE) antibodies were applied to assess the presence of these molecules on CD4 and CD8 T cells in PB. Briefly, preparation for flow cytometry was as follows: to each cytometric tube, 100 μL of LNs aspirate or PB and 4 μL of specific monoclonal antibodies were added. After 15 min of incubation in the dark, at room temperature, erythrocytes were lysed with lysing solution for 10 min and washed with 2% newborn calf serum in physiological buffer solutions (PBS). The cells were subsequently fixed in PBS. The samples were processed by the FACS Canto II flow cytometer (BD, USA). Geometric mean fluorescence (GMF) intensity of PD-L1, CD47, CD73, Fas, and FasL on MTCs and CSCs was measured. The analysis was performed using BD FACSDiva™ Software (BD, USA)
For the statistical analysis, the Mann–Whitney U-test was performed to: compare the differences between patients with confirmed metastases in LNs and patients without metastases, and compare the differences between the expression of immunomodulatory molecules on CSCs and MTCs. Correlation analyses were performed by calculating the Pearson r coefficient. Differences were considered statistically significant when p < 0.05. All analyses were performed using Prism (Version 5, GraphPad Software, La Jolla, CA, USA).