Abstract
Background/Aim: Lung cancer is among the cancers with the highest mortality rates worldwide. Biomarkers associated with it are critically important for diagnosis, evaluation of clinicopathological features, and the development of treatment strategies.
Patients and Methods: In this study, a total of 32 NSCLC primary tumor tissues and 32 matching 32 normal tissues were analyzed with Western Blot for CD70, CD27, CD3, and FOXP3; positive and negative results were evaluated according to a quantitative analysis of expression with ImageJ. Furthermore, the levels of sCD27 in the serum samples of 30 NSCLC patients and 32 healthy controls were investigated by ELISA.
Results: CD70 expression was observed in 5/32 (15.63%) NSCLC tumors, CD27 in 24/32 (75%), CD3 in 28/32 (87.5%), and FOXP3 in 14/32 (43.75%) tumor tissues, all significantly differing from normal tissues (p<0.0001). The mean serum sCD27 level in NSCLC patients (117.29±38.18 U/ml) was considerably higher than in controls (p<0.0001). Positive CD70 expression in tumors was significantly correlated with higher serum sCD27 levels compared to CD70-negative tumors (x0.007). No significant association was found between overall survival and sCD27 status (p=0.779).
Conclusion: The findings suggest that the CD27/CD70 pathway is a potential diagnostic and therapeutic target in NSCLC. Serum sCD27 levels may serve as a diagnostic biomarker. CD70, CD27, CD3 and FOXP3 quantifications show that CD70 is expressed in NSCLC, and related molecular mechanisms support previous findings. However, further studies are required with larger cohorts in NSCLC.
Introduction
Lung cancer is the second most frequently diagnosed cancer and the leading cause of cancer-related mortality worldwide as of 2020 (1). It is classified into non-small cell lung cancer (NSCLC), accounting for 85% of cases, and small cell lung cancer (SCLC), making up the remaining 15% (2). NSCLC is often diagnosed at an advanced stage, necessitating varied treatment approaches depending on tumor progression (2-4). While surgery, radiotherapy and chemotherapy are the primary interventions, there is an increasing need for novel therapeutic strategies to improve patient outcomes (2-4). Immunotherapy has emerged as a promising alternative for NSCLC screening, diagnosis, and treatment (3-6).
The tumor microenvironment (TME) consists of immune and inflammatory cells, fibroblasts, extracellular matrix (ECM), and blood vessels, playing a crucial role in cancer progression and drug resistance (6, 7). Understanding TME is essential for developing effective diagnostic and therapeutic strategies (7, 8). Notably, immune checkpoint inhibitors targeting molecules such as programmed cell death protein 1 and its ligand (PD-1/PD-L1) and cytotoxic T-lymphocyte antigen 4 and its ligand (CTLA-4/CD28) have revolutionized NSCLC treatment (2, 5-8).
CD27, a member of the tumor necrosis factor receptor superfamily (TNFRSF), plays a vital role in immune cell activation, with its ligand CD70 being transiently expressed on activated immune cells and various tumors, including clear cell renal carcinoma (ccRCC) (9-13). CD70 is also under investigation in glioblastoma, ovarian, colorectal, and NSCLC for its diagnostic and therapeutic potential (3, 14-17). CD27/CD70 interaction influences T cell priming and differentiation, leading to soluble CD27 (sCD27) release into the circulation upon cleavage (3, 9, 13). Aberrant CD70 expression in tumors can facilitate immune evasion by promoting regulatory T cell (Treg) accumulation, contributing to immunosuppression and tumor progression (3, 7, 9, 13, 18). Consequently, monoclonal antibodies targeting CD70 are being developed for therapeutic applications (9, 13, 18).
Given the potential of CD70 as a biomarker and therapeutic target, this study quantitatively analyzed its protein expression in NSCLC TME along with its receptor CD27. Additionally, the presence of T cells and Tregs was assessed through CD3 and FOXP3 protein expression in tumors and normal tissues.
Materials and Methods
Study population and sample collection. In this study, a total of 32 surgically removed tissue samples taken from NSCLC patients were used for protein expression analysis, and 30 serum samples taken from NSCLC patients and 32 serum samples taken from healthy controls were used for ELISA. The study was approved by the Clinical Research Ethics Committee of Istanbul Medicine Faculty (No: 2017/884) for tissue and serum sample collection and informed consent was signed and obtained for all patients. While a total of 46 NSCLC patients with an average age [± standard deviation (SD)] of 58.72±9.33 (9 females and 37 males) were analyzed in this study. A total of 32 healthy controls samples (7 females and 25 males, average age of 50.87±1.97) were only used for ELISA. Clinicopathological characteristics of the patients were also collected and evaluated.
Tissue sample preparation and protein isolation. A total of 32 surgically removed primary tumor tissue and normal tissue samples taken from NSCLC patients were stored at −80°C before the experiment. TRIzol® Reagent (TRIzol® Reagent, Ambion by Life Technologies, Carlsbad, CA, USA, Cat. no. 15596-026) was used according to the manufacturer’s instructions for 32 NSCLC primary tumor tissue and normal tissue samples. 50-100 mg of frozen tissue samples taken from −80°C and thawed on ice were homogenized with 1 ml of TRIzol® Reagent at room temperature (RT). Before phase separation, all homogenized samples were kept at −80°C. The final protein products were resuspended and stored in 1% SDS at 20°C for further usage.
Detection of protein expression by western blot. Isolated protein samples were used for western blot to detect CD70, CD27, CD3, and FOXP3 protein expression in tissue samples. Protein concentrations were determined using the SMART BCA Protein Assay Kit (Cat. No. 21071, iNtRON Biotechnology, Boston, MA, USA) according to the manufacturer’s instructions. Each protein sample (200 μg) was loaded onto a polyacrylamide gel for SDS-PAGE, and protein blotting was carried out using the Mini-PROTEAN® Tetra System and Mini Trans-Blot® Cell System (Bio-Rad, Hercules, CA, USA). Gel electrophoresis run was performed at 50V for 15 min followed by 150 V for 1 h. The wet transfer system was employed for blotting in the cold room (4°C) at 100 V, 350 mA for 1 h using Immun-Blot® PVDF Membrane Sandwiches (Bio-Rad, Cat. no. 162-0219). PVDF membranes were blocked with 5% skimmed milk in Tris-buffer saline with Tween 20 (TBST) buffer for 1 h at RT on a shaker. Primary antibody incubations were carried out overnight at 4°C on a shaker with antibodies prepared in blocking solution (5% skimmed milk in TBST) according to the manufacturer’s instructions. Anti-CD70 Antibody (Abcam, Cambridge, UK, ab175389, 1:2,000), anti-beta-actin Antibody (Abcam, ab119716, 1:5,000), anti-CD27 Antibody (Abcam, ab70103, 1 μg/1 ml), anti-FOXP3 Antibody (Abcam, ab70103, 1:500), and anti-CD3 Antibody (Abcam, ab70103, 0.5 μg/1 ml) were used. The secondary antibody incubation was performed with a rabbit IgG - Fc (HRP) antibody (Abcam, ab97069) prepared in blocking solution (5% skimmed milk in TBST, 1:5,000) according to the manufacturer’s instruction at RT for 1 h. Clarity™ Western ECL Substrate Kit (Bio-Rad, Cat. no. 1705060) was used as a substrate for chemiluminescent detection prepared according to the manufacturer’s instructions. Chemiluminescent signals were captured using a CCD camera-based imager.
Quantification of protein expression by ImageJ. ImageJ 1.52a software was used to calculate the band intensities of each blot in this study. Before quantification, blot images were converted to 8-bit format and uncalibrated optical density was applied using ImageJ software. Background subtraction was applied using the rolling ball radius approach, followed by a one-by-one selection of each band and encircled by a rectangle called Region of Interest (ROI) selection with the “Gels” function. Following the recording of histograms, the peak area of each band was measured, and data was recorded as arbitrary area values. To get an image with clear bands, a well-documented approach (19) for quantifying blots known as “background subtraction” was used as described. For each band, three independent readings were taken, and a relative band density was determined by normalizing the computed data using beta-actin quantification. GraphPad Prism Version 8.2.1 for Windows was then used to perform statistical analysis on the final corrected density. Results were represented as relative densities in arbitrary units (a.u).
Enzyme-linked immunosorbent assay (ELISA) for determination of sCD27 levels in patient serum. Blood serum was obtained from NSCLC patients and healthy donors by centrifugation and then stored at −80°C. Human sCD27 INSTANT ELISATM Kit (ThermoFisher Scientific, Horsham, UK) was employed for quantitative detection of human soluble CD27 (sCD27) levels in serum samples according to the manufacturer’s instructions.
Statistical analysis. Man-Whitney U test, Kaplan-Meier, 2-way ANOVA and ROC analysis were performed for statistical analysis using SPSS Version 17.0 (IBM Corp., Armonk, NY, USA). Graphs illustrated using GraphPad Prism Version 8.2.1 (GraphPad Software, Boston, MA, USA).
Results
No significant correlation was found for age and gender parameters of patients and healthy controls in the study group. A total of 32 NSCLC primary tumor tissues and matching 32 normal tissues were analyzed with western blot for CD70, CD27, CD3, and FOXP3 proteins; positive and negative results were determined according to the quantitative analysis of bands on the blots with ImageJ and GraphPad Prism as described earlier. A total of 5/32 samples (15.63%) were found to be expressing the CD70 protein in NSCLC primary tumor tissues, 24/32 samples (75%) were positive for CD27, 28/32 samples (87.5%) were positive for CD3, and 14/32 samples (43.75%) were positive for FOXP3.
The quantified blot expression analysis is represented with bar graphs. Tumor and matched nt-tissue samples from each of the 32 NSCLC patients were analyzed, and quantification was carried out. GraphPad Prism 8.2.1 was used to examine the quantitative data, and Student’s t-test was used to determine significance. Statistical analysis revealed that for all patients, the differences in tumor and normal tissue expression were statistically significant (p <0.01) (Figure 1). A total of 32 NSCLC patient samples were named from P1 to P32. Each patient includes 2 samples as primary tumor tissue and normal tissue, and naming was done according to this. While only 5 primary tumor tissues were found to be positive for CD70, there was no band quantified on any normal tissues for CD70 (Figure 1a). Twenty-four bands were quantified as positive for CD27 on tumor tissues; on the other hand, a total of 13 normal tissues were expressing CD27 (Figure 1b). Twenty-eight of tumor tissues were found to be expressing CD3, while only 8 normal tissues were positive for CD3 (Figure 1c). Lastly, 14 tumor tissues were quantified as positive for FOXP3 and a total of 5 normal tissues were expressing FOXP3 (Figure 1d).
Bar graphs representing the expression levels of CD70, CD27, CD3, and FOXP3 proteins in tumor and normal tissue samples from patients with non-small cell lung cancer (NSCLC). Protein expression levels are presented in arbitrary units (a.u.). Each bar reflects the mean of three technical replicates for each protein in either tumor or adjacent normal tissue. (a) CD70, (b) CD27, (c) CD3, and (d) FOXP3 protein expression levels were significantly elevated in tumor tissues compared to corresponding healthy tissues. Statistical analysis was conducted using a two-way ANOVA, which assessed: (i) the main effect of tissue type (tumor vs. healthy; row Factor), (ii) the main effect of protein type (CD70, CD27, CD3, FOXP3; column Factor), and (iii) the interaction between tissue and protein expression (interaction). All three effects were statistically significant with p<0.0001. The figure illustrates averaged expression values per tissue and protein, and the statistical significance of differences between tumor and control groups is indicated by horizontal lines with asterisks. Asterisks denote p<0.0001. Error bars represent the standard error of the mean (SEM). p<0.05 was considered statistically significant.
The patients’ clinicopathological features were then correlated with the levels of expression of CD70, CD27, CD3, and FOXP3 molecules. There was no significant relationship identified the between expression status and the clinicopathological features. To assess the sCD27 serum levels, 30 NSCLC patients and 32 healthy controls had their blood samples taken, and serum samples were used for ELISA. The mean level (mean±SD) of sCD27 in NSCLC patients was found to be 117.29±38.18 U/ml (range=50.24-236.49 U/ml), and healthy controls were found to be 82.38±21.99 U/ml. The difference in the mean level of sCD27 level in NSCLC patients and healthy controls was statistically significant (p<0.0001). Also, ROC analysis revealed that the cut-off value for sCD27 was 88.91 U/ml (Figure 2). Furthermore, sensitivity was determined to be 90% and specificity was shown to be 72%. The ELISA findings demonstrated excellent sensitivity and specificity, indicating that the test properly detects individuals with high sCD27 levels with high significance (p<0.0001). In addition, the measured sCD27 serum levels were correlated with the clinicopathological features of the patients, but no association was established between the sCD27 levels and the clinicopathological parameters. Finally, sCD27 serum levels were correlated with the expression status of CD70, CD27, CD3, and FOXP3 for matched 16 NSCLC patient samples (Table I). High sCD27 serum level was observed to be associated with CD70 positivity on tumor microenvironment in NSCLC patients (p =0.007). The mean level (mean±SD) of sCD27 levels in a total of 5 NSCLC patients with CD70 expression on TME was found to be 155.96±54.48 U/ml, while the levels of rest of the 11 NSCLC patients with no CD70 expression on TME were 95.79±19.87 U/ml. Although CD27 and FOXP3 did not correlate with sCD27 levels, the CD3 association was showing a trend towards statistical significance (p =0.091). There was no statistically significant difference between overall survival and sCD27 status in the NSCLC patients (p=0.779) (Figure 3).
Correlation between sCD27 serum levels and expression status of CD70, CD27, CD3, and FOXP3 in the tumor microenvironment of NSCLC patients as determined by the Mann-Whitney U test.
Sensitivity and specificity were determined by ROC Curve analysis for sCD27 levels of NSCLC patients and healthy controls. sCD27 levels were counted as positive for the values above 88.91 U/ml. The patient group has a sensitivity of 90% and specificity of 72% (p<0.0001 as determined by DeLong test).
Kaplan-Meier curve shows the association between the positive (>88.91 U/ml) and negative (≤88.91 U/ml) sCD27 status and overall survival (p=0.779).
Discussion
Potential biomarkers and diagnostic tools have been extensively investigated across various cancer types, with immune checkpoint molecules gaining increased attention in recent years. It is critical to have precise biomarkers that enable lung cancer detection and diagnosis in order to properly diagnose the tumor, evaluate patient features, and develop potential therapeutics (2). CD70 has been demonstrated to be a rather promising target molecule as a NSCLC diagnostic. Additionally, the CD27/CD70 pathway has emerged as a promising therapeutic target and warrants greater emphasis in the context of NSCLC. Therefore, this study explored the molecular mechanisms associated with the CD27/CD70 axis within the tumor microenvironment (TME) of NSCLC.
CD27 is expressed on tumor-infiltrating lymphocytes located in the TME of solid tumors, promoting priming and memory development of T cells via the CD27/CD70 interaction (3, 9, 20). It has been found that the CD27/CD70 interaction is necessary for the proper immune response against the tumor, although tumors can revert this into their favor and diminish T cell function by several mechanisms (20, 21). The transient CD70 expression is tightly controlled on activated lymphocytes; however, it has been shown that several hematologic malignancies and solid tumors aberrantly express CD70 which finally leads to continuous interaction of CD27/CD70 on the TME (11, 22, 23). Once this persistent signaling has started, it eventually causes T cell exhaustion and Treg accumulation into the TME (21, 23). Due to the immunosuppressive function of regulatory T cells (Tregs) and the high expression of CD70 on tumor cells, an accumulation of FOXP3-expressing Tregs has been reported in the tumor microenvironment (TME) of NSCLC (21, 24). Therefore, the quantitative expression status of CD70, CD27, CD3, and FOXP3 proteins were examined in 32 NSCLC tumors and normal tissues in this study. sCD27 levels were also investigated in order to analyze the association of soluble levels of CD27 in NSCLC patients with CD70, CD27, CD3, and FOXP3 expression in TME. According to several studies, sCD27 levels have also been associated with patients’ clinicopathological characteristics (25). In addition, the diagnostic potential of these targets has been evaluated in the context of NSCLC, especially CD70 is a promising target for cancer immunotherapy because of the absence of its normal tissue expression (18, 21). ccRCC is one of the cancer types where the strongest CD70 overexpression was detected, and glioblastoma, ovarian cancer, colorectal cancer, and thyroid cancer have been also found to be strongly positive for CD70 (11, 14, 16, 17, 23, 25). However, further investigation is necessary for NSCLC. CD70 expression was quantified in only 5/32 (15.63%) of NSCLC tissues in this study, and none of the normal tissues exhibited CD70 positivity after quantification, as predicted. A high amount of positivity was not detected for NSCLC with 15.63% in this study, although larger cohorts are required for future investigation. Furthermore, Jacobs et al. (17) discovered that only 16.3% of the tissues examined by immunohistochemistry (IHC) in NSCLC were positive for CD70. Thus, our results are correlated with previous results, even though their study also includes a small number of cohorts (17, 26). Furthermore, no association was identified between histological types or any other clinicopathological characteristics of patients and CD70 positivity.
Quantification of CD3 and FOXP3 expression revealed that similar tumor tissues were expressing both proteins together in TME. In addition, all CD70 expressing tissues were expressing both CD3 and FOXP3 as well (Figure 1a, c, d). The fact that tissues with CD70 expression also have FOXP3 and CD3 expression supports earlier findings which show that CD70 expression in TME is associated with FOXP3+ Treg and CD3+ T cell accumulations (17, 25, 27, 28). However, additional samples are needed to get to a valid conclusion about their connection with TME, since only 5 tissues out of 32 have CD70 positivity. The current technique does not allow the detection of the expression location, although quantification of the expression validates the presence of the proteins in NSCLC TME. In this study, 22/32 patients had identical CD27 and CD3 expression according to the expression analysis. Ruf et al. discovered that CD27+ TILs in the TME of ccRCC tissues are actually CD3+ T cells, which supports our findings (25).
The presence of sCD27 was identified in the serum samples of 30 NSCLC patients and 32 healthy controls. The level of sCD27 was shown to be statistically significant for the patient group with high sensitivity (90%), indicating that sCD27 may be used as a diagnostic biomarker for NSCLC diagnosis. However, no significant association was discovered between the patients’ clinicopathological characteristics and sCD27 levels. As a result, sCD27 can predict the presence of NSCLC in a diagnostic manner; however, patient prognosis cannot be predicted according to the level of sCD27 in the serum.
In NSCLC tissues, the expression of CD70, CD27, CD3, and FOXP3 was compared to the level of sCD27. CD70 positivity on tumor tissues was shown to be strongly correlated with the level of sCD27 in serum when compared to CD70 negativity, indicating that CD70/CD27 interaction in TME induces CD27 cleavage and an increase in sCD27 in patient sera. There was no statistically significant relationship between CD27 and FOXP3 positivity and sCD27 levels. Furthermore, the level of sCD27 demonstrated a marginally significant association with CD3 positivity, supporting earlier research that indicated that increased CD3 accumulation in the TME is linked to patients with higher levels of sCD27 in the patient sera (11, 22, 23, 29).
Study limitations. Firstly, the relatively small sample size may limit the statistical power and the generalizability of the results. Secondly, CD70 expression was observed in only 5 out of 32 tumor samples, which may not fully represent the broader tumor population. Future studies with larger cohorts are warranted to validate these findings and further explore the role of CD70 in tumor biology.
Conclusion
In conclusion, the current findings reveal that the CD27/CD70 pathway is a potential diagnostic and therapeutic candidate for NSCLC. sCD27 levels can be utilized for diagnostic purposes. CD70, CD27, CD3, and FOXP3 quantifications showed that NSCLC can express CD70, and related molecular mechanisms support previous findings. However, further studies are required with larger cohorts in NSCLC.
Acknowledgements
We would like to express our sincere appreciation to Prof. Dr. Bulent Ozpolat (Houston Methodist Research Institute) for his valuable contributions to the evaluation of the data. We would also like to express our gratitude to Dilara Sonmez Zor, Ph.D., and Lecturer Mehmet Tolgahan Hakan, Ph.D., for their effort and kindness during the project. Also, we would like to thank Gunes Ozen Eroglu, Ph.D., for providing a positive control cell line for the experiments.
Footnotes
Authors’ Contributions
Ilhan Yaylim: Project administration, conceptualization, supervision, designing experiments, provided critical feedback and helped shape the research, analysis and manuscript. Merve Saide Uzunoglu: Investigation, designing and performing experiments, analyzing data and writing the manuscript. Aylin Seher Uzunoglu: Processed the experimental data, performed the analysis, drafted the manuscript and designed the figures. Akif Tuma: Taking responsibility in patient follow-up, collection of relevant biological materials, data management and reporting, providing clinical samples, analyzing data, interpreting the results and worked on the manuscript. Volkan Kara: Providing clinical samples, data management and reporting. Ozlem Kucukhuseyin, Cem Horozoglu: Conceptualization, designing experiments, provided critical feedback. All Authors critically revised the manuscript, approved the final version to be published, and agreed to be accountable for all aspects of the work.
Supplementary Material
Available at: https://doi.org/10.6084/m9.figshare.29260673.v1
Conflicts of Interest
The Authors declare that they have no conflicts of interest.
Funding
This Project was supported by the Istanbul University Scientific Research Projects Committee Project No: TYL-2018-29527.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received March 9, 2025.
- Revision received May 23, 2025.
- Accepted June 9, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
Jump to section
Related Articles
Cited By...
- No citing articles found.