NSCLC-associated specimens were obtained from 92 patients with NSCLC (50 men; 42 women; mean age 41.2 years, age range 28–76 years) undergoing surgery at the Department of Thoracic Surgery in The Second Hospital of Tianjin Medical University between October 2014 and June 2018. Two experienced pathologists at the Department of Pathology in The Second Hospital of Tianjin Medical University confirmed the diagnosis independently (14). The stained methods have been provided in IHC sections. None of patients had received chemotherapy and radiotherapy before the operation. The patient information, including the general characteristics, diagnosis, treatment strategy and the survival time, were obtained from the medical records or outpatient follow-up records of the patients every 3 months. All of the operative tissues [normal lung (distance to tumor tissues, 2 cm) and tumor tissues] were harvested during resection. Subsequently the samples were immediately frozen with liquid nitrogen and stored until further use. The aforementioned procedures were approved by the Research Ethics Committee of Tianjin Medical University. All patients provided written informed consent in accordance with the Declaration of Helsinki.

The human H460, H1299 and A549 NSCLC cell lines, and the PC6 SCLC cell line, were purchased from the Institute of Basic Medical Sciences Peking Union Medical College. The PC6 cells were used as a control. The 4 tumor cell lines were authenticated using cellular morphology and cultured in DMEM (Thermo Fisher Scientific, Inc.) supplemented with 10% FBS (Thermo Fisher Scientific, Inc.) and penicillin/streptomycin (1%) at 37°C in a humidified incubator with 5% CO₂. H460 and A549 cell lines in the following experiments were divided into four groups: i) Pem combined with ¹²⁵I (Pem + ¹²⁵I, 10 µg/µl); ii) Pem 10 µg/µl; iii) ¹²⁵I, 10 µg/µl; and iv) DMSO, 10 µg/µl.

Pem (5 ml; 20 mg; molecular weight 534.6) was purchased commercially from Merck KGaA. ¹²⁵I radioactive seeds with 22.4–29.6 MBq/particle were purchased from Shanghai GMS Pharmaceutical Co. Ltd. Both reagents were stored at −20°C and a small amount (10 µg/µl) was used for subsequent experiments.

IHC staining procedures were performed according to the manufacturer's instructions. Tissues were dehydrated by using ethanol gradient (70, 80 and 95%; 5 min each), followed by incubation with 100% ethanol three times for 5 min. The paraffin-embedded block was sliced into 5–8 µm thick sections using a microtome and washed in a 40°C water bath containing distilled water for 15 min. Section were fixed in 4% PFA (Image-iT™ Fixation/Permeabilization Kit; Invitrogen; Thermo Fisher Scientific, Inc.) at 4°C overnight. The sections were heated at 75°C for 30 min and 5% normal FBS (Thermo Fisher Scientific, Inc.) to reduce unspecific background staining. Briefly, to investigate the expression of PD1L1 in lung cancer tissue and investigate the association with tumor proliferation, the prepared sections were stained with anti-PD1L1 antibody (cat. no. CST51296; 1:100; Cell Signaling Technology, Inc.) and anti-Ki-67 antibody (cat. no. AB15580; 1:100; Abcam) at 4°C overnight. The validity of these antibodies has been confirmed by previous studies (15,16). Subsequently, the samples were washed with PBS and incubated with the goat to rabbit secondary antibodies (Thermo Fisher Scientific, Inc.; cat. no. A11034; 1:500) for 1 h at room temperature. The ratio of PD1L1-positive cells and Ki-67-positive cells was calculated by counting the average number of stained cells in 500 nuclei in four random light high-power fields (magnification 400X (17). The average IHC score of PD1LI in all samples was 6. Histopathological diagnoses were confirmed by two pathologists at The Second Hospital of Tianjin Medical University based on the WHO classification of thorax tumors.

To determine cell proliferation, MTT assays were strictly performed according to our previous work (18). The H460 cells (1,000 cells/well) were seeded into 96-well plates for 24 h (37°C, 5% CO₂) and subsequently treated with different compounds (10 µg/µl) based on the aforementioned groups for an additional 24 h. Cell viability was assessed using the MTT assay. Briefly, 0.5 mg MTT reagent was added to each well for 4 h of incubation before the end of Pem or ¹²⁵I treatment. After removing the supernatant, 100 µl DMSO was added to each well and incubated for 10 min. The purple mixture in each well was then measured at 490 nm using the Polarstar Optima microplate reader from BMG Labtech GmbH. The inhibition rate was analyzed using the following formula: [(Control D490-Experimental D490)/Control D490] ×100%.

FCM analysis was performed according to the manufacture's instructions. Briefly, tumor cells were diluted to 1×10⁵/ml and treated with the aforementioned reagents, in the four treatment groups for 24 h. The cells were dissociated in 1X binding buffer and subsequently stained with Annexin V-FITC (Biolegend, Inc.) and propidium iodide (PI) in the dark for 20 min at room temperature and analyzed using FCM (Becton, Dickinson and Company) cell sorting. The apoptosis rate was calculated as the percentage of the Annexin V-FITC⁺/PI⁺ cells from the total cells. Additionally, the dissociated cells were incubated in 95% cold alcohol overnight at 4°C and then stained with PI (50 ug/ml) supplemented with Triton X-100 in the dark for 30 min at 4°C. In total, 1×10⁴ cells were used as the gate region and analyzed using FCM cell sorting to examine the cell cycle. The dead cells were calculated as the proportion of PI⁺ cells. The data is presented as the mean ± standard deviation from three independent experiments.

ELISA was performed using the matrix metalloproteinase (MMP)2 and MMP9 ELISA kits (Boster Biology Technnology; cat. nos. EK1488 and EK1107) according to the manufacturer's instructions. Briefly, 100 µl supernatant from H460 and A549HE cells cultured with the aforementioned reagents for 8 h was added into duplicate wells and incubated at 37°C for 90 min. After washing three times, 100 µl indicating antibodies from the ELISA kits were added and incubated at 37°C for 60 min. After the color reaction step with 3,3′,5,5′-tetramethylbenzidine substrate for 15 min at 37°C, absorbency was detected at 450 nm using a microplate reader (SpectraMax; Molecular Devices, LLC). The data is presented as the mean ± standard deviation from three independent experiments.

Cells from all 4 cell lines were lysed with ice-cold buffer (50 mM Tris. Cl pH 6.8, 15 mM NaCl, 5 mM EDTA, 0.5% NP-40 and 1 mM PMSF) supplemented with 0.1% Triton X-100 and 20 mM Tris-HCl (pH 7.4), and protease-inhibitor cocktail. The protein concentration was determined with Coomassie Brilliant Blue (Thermo Fisher Scientific, Inc.; cat. no. 23200) using a microplate spectrophotometer (Infinite M200 PRO; Tecan Group, Ltd.).

The total protein (50 µg) was separated with 10% SDS-PAGE gel and the protein was transferred on to PVDF membranes. The membranes were subsequently incubated against the Bax (cat. no. AB182733; Abcam) and Bcl-2 (cat. no. AB32124; Abcam) antibodies overnight at 4°C. Subsequently, the membranes were probed with secondary antibodies (cat. no. GL1538; GuideChem; 1:500) for 1 h at room temperature. β-actin was utilized as the loading control. The primary antibodies were diluted to 1:500 and the β-actin antibody was diluted to 1:3,000. The immunoblotting results were visualized using TMB (cat. no. PR1210; Beijing Solarbio Science & Technology Co., Ltd.).

RNA was isolated from tumor tissues and cells from all 4 cell lines following the treatment of the aforementioned regents using TRIzol^® (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. cDNA was synthesized using iScript reverse transcription system (Invitrogen; Thermo Fisher Scientific, Inc.) and 2 µg RNA. Real-time RT-PCR conditions were 95.0°C for 5 min, 35 cycles of 95.0°C for 30 sec and 60.0°C for 30 sec. The mRNA expression levels of PD1L1, p21, p27 and ADAM metallopeptidase domain 17 (ADAM17) were determined using the iQ SYBR Green Supermix kit (Bio-Rad Laboratories, Inc.) and the Bio-Rad iQ5 Multicolor Real-Time PCR detection system and normalized with GAPDH mRNA. The following conditions were used for the qPCR: Initial denaturation at 95.0°C for 5 min, then 35 cycles of 95.0°C for 30 sec and 60.0°C for 30 sec. The 2^−∆∆Ct method (19) was performed to calculate the fold change and all the experiments were conducted three times. The sequences used for the primers are provided in Table SI.

Student's t-test and χ² test were used to determine the statistical difference between the reagent treatment group and the control group. Data were presented as the means ± standard deviation. Data were analyzed by Student's t-test and Mann-Whitney test using SPSS v19.0 (IBM Corp.) and GraphPad Prism v7.0 (GraphPad Software, Inc). An unpaired t-test was used for comparison between two groups, and comparison of mean values between multiple groups was evaluated by one-way ANOVA followed by Student-Newman-Keuls post hoc test. The functional experiments were performed at least three times. The correlation between PD1L1 expression and Ki-67 labeling was determined using Pearson's correlation. The clinical risk factors were analyzed using logistic regression. Survival analysis was performed using Kaplan-Meier curves. The Mantel-Cox log-rank test was used to assess the significant differences between survival curves. P<0.05 was considered to indicate a statistically significant difference.

A total of 92 NSCLC cases were included in the present study: 71 (77.17%) were squamous cell carcinoma, 16 (17.39%) were adenocarcinoma and 5 (5.43%) were large cell carcinoma (Table I). The expression of PD1L1 was examined using IHC. PD1L1 was expressed in human NSCLC samples and was located on the surface of tumor cells (Fig. 1A). As presented in Table I, the IHC score of PD1L1 was <6 in 33 (35.87%) cases and >6 in 59 (64.13%) samples. he Ki-67 index of the 92 NSCLC cases was determined and the samples were subsequently divided into three groups depending on the percentage of cells stained positive. Ki-67 labeling was identified as low percentage (<10.0%) in 28 (30.43%) cases, moderate percentage (10.0–20.0%) in 34 (36.96%) cases and high percentage (>20.0%) in 30 (32.61%) cases. Most of the cases had moderate percentage expression. Similarly, most of the patients (49; 53.26%) were at the second risk stage (Table I). Additionally, the association of PD1L1 expression with Ki-67 index was evaluated by Pearson's correlation, which revealed positive correlation (Fig. 1A). These data suggests that PD1L1 expression was associated with proliferation of NSCLC.

In the present study, the median follow-up time was 35.8 months (range 3.0–59.0 months). Among them, 39 (42.39%) patients were alive and 53 (57.61%) patients died as a result of NSCLC during the follow-up periods. In addition, among the patients (dead and alive) with PD1L1 expression, progression-free survival (PFS) and overall survival (OS) were calculated using Kaplan-Meier analysis. Patients with lower levels of PD1L1 expression (<6.0, the average score of all cases) had a significantly longer PFS compared with those with higher expression levels (Fig. 1B). Similarly, patients with lower expression had a longer survival (Fig. 1C). As shown in Table I, the parameters of age, risk stage, PD1L1 score, ratio of Ki-67 labeling and pathological subtypes also affected the OS as evidenced from the logistic regression analysis. The aforementioned data demonstrated that PD1L1 could be a prognostic indicator for patients with NSCLC.

Following the association of PD1L1 with NSCLC progression, the role of PD1L1 in tumor proliferation was subsequently explored. For this purpose, PD1L1 mRNA expression was investigated in H1299, H460 and A549 NSCLC cell lines and the PC6 SCLC cell line, using RT-qPCR to select the appropriate cell model for further experimentation. As presented in Fig. 2A, among the four cell-lines, the relative mRNA expression of PD1L1 was significantly higher in H460 and A549 cells, but there was low expression in the H1299 cell line (~0.1-fold) and PC6 cell line (~0.22-fold), therefore the H460 and A549 cells would be utilized in the subsequent experiments. The expression of PD1L1 in fresh human NSCLC tissues was determined, and it was found to be significantly higher compared with that in normal tissue samples (Fig. 2B). To detect the anticancer effect of Pem and ¹²⁵I on NSCLC cells, the MTT assay was performed using the aforementioned experimental groups, including the combination of Pem with ¹²⁵I, Pem, ¹²⁵I and DMSO as the control. As shown in Fig. 2C, the proliferation curves revealed that the combination of 2.0 µmol/l Pem and radioactive ¹²⁵I seeds significantly reduced H460 cells (left) and A549 (right) growth, compared with any of the single treatment groups. Although the inhibitory effects were observed in the Pem or ¹²⁵I radiation single treatment groups, the suppression of proliferation was strongest in the combination treatment group, supporting the hypothesis that the combination of the two strategies could reduce tumor cell growth. To detect the effects of Pem+¹²⁵I on the migration ability of tumor cells, H460 and A549 cells were generated based on the above subgroups for 8 h and then the concentrations of secreted MMP2 and MMP9 proteins were tested in the supernatant of NSCLC cell cultures by ELISA. As demonstrated in Fig. 2D and E, MMP2 secretion was suppressed more efficiently by Pem combined with ¹²⁵I treatment group compared with that in the single treatment groups of Pem and ¹²⁵I, and the control, for both cell lines. The secretion of MMP9 was also similarly inhibited by the combination treatment group compared with that in the single treatment groups in both cell lines (Fig. 2D and E). These data suggests that the combination of Pem and ¹²⁵I played a critical role in repressing the tumor proliferation and aggressiveness.

Apoptosis in H460 and A549 cells treated with the aforementioned reagents was determined using FCM analysis. As presented in Fig. 3A, the combination of Pem and ¹²⁵I could induce tumor cell apoptosis up to ~65%, whereas the single Pem or ¹²⁵I radiation treatment groups only induced apoptosis to <40%. According to cell proliferation mediated by the cell cycle, FCM was continuously performed to further detect the effect of the combination treatment on cell cycle by measuring the binding activity of PI. As shown in Fig. 3B, the combination treatment resulted in more H460 cells arrested in in S phase without ability to enter into the proliferation cycle and any of the single treatment analogously inhibited the blocking ability about 20%. Thus, combination with Pem and ¹²⁵I showed the strong anti-tumor activity via suppressing the proliferation and mobility as well as promoting apoptosis.

Following the observation that the combination of Pem and ¹²⁵I was more effective in the NSCLC cell models compared with the single treatment groups, the underlying mechanisms involved in the anti-tumor effects were subsequently investigated. Based on the striking ADAM17 phenotype, the protein and mRNA expression levels of key proteins associated with apoptosis and proliferation, such as Bcl-2, Bax, p21 and p27 (20,21), in H460 cells treated with the aforementioned reagents were detected using western blot analysis and RT-qPCR. As presented in Fig. 4A, markedly decreased levels of Bcl-2 were observed in H460 cells following combination treatment compared with that in the single treatment groups. In addition, this combination also markedly upregulated Bax expression to induce the apoptosis in contrast with the single Pem or ¹²⁵I treatment groups. Furthermore, the mRNA expression of p21 and p27 was determined using RT-qPCR, which are considered as regulators of the cell cycle (22,23). As shown in Fig. 4B, both p21 and p27 mRNA levels were significantly elevated in H460 cells (~40%) following the combination treatment, supporting the phenomenon that most of the tumor cells were arrested in S phase. Recently, ADAM17 was discovered as a crucial mediator of resistance to immunotherapy, therefore (24) the mRNA expression of ADAM17 was investigated. There was a significant reduction in ADAM17 levels in the combination treatment group compared with the control group (Fig. 4C), suggesting that the combination could modify the sensitivity of H460 cells to the targeted therapy and radiation. Therefore, these data demonstrate that the combination of Pem and ¹²⁵I treatment suppressed NSCLC cell growth and induced the apoptosis through ADAM17 expression.