The MKN-45 cell line (human gastric cancer cell line) was purchased from the Iranian Biological Resource Center (Tehran, Iran). Epstein-Barr virus (EBV)–transformed lymphoblastoid cell line (LCL) as feeder cells was maintained in Roswell Park Memorial Institute-1640 (RPMI-1640) (Gibco, United States); media were supplemented with 10% FBS (Gibco, United States), 1% penicillin-streptomycin, and 1% L-glutamine.

NK cells were isolated from healthy donors’ buffy coats with Ficoll Paque Premium (GE Healthcare’s, United States) gradient centrifugation. According to the manufacturer’s instructions, NK cells were collected by negative selection using a human NK cell isolation kit (Miltenyi Biotech, Germany). The purity of NK cells was assessed by flow cytometry analysis of CD3⁻ and CD56⁺ markers (FACSCalibur Becton Dickinson, United States). EBV-LCL was applied for optimal NK cell expansion. EBV-LCL cell line was established by culturing peripheral blood mononuclear cells (PBMCs) in the presence of 100 μg/ml cyclosporin A with EBV supernatant harvested from the cell line B95-8 (Iranian Biological Research Center) (Igarashi, Wynberg et al., 2004). The procedure is based on an existing expansion protocol for in vivo studies (Berg, Lundqvist et al., 2009) that utilizes 100 Gy-irradiated EBV-LCL as feeder cells (at a ratio of 1:10) trigger NK cell proliferation and highly activated NK cells. In the first 5 days, NK cell colonies are formed, and every 3 days, a fresh medium enriched by IL-2 (500 IU/ml) (Miltenyi Biotech, Germany) is added to the cells for 21 days.

The cytotoxic effects of activated NK cells as an effector cell (E), whether alone and in combination with anti-PD-1, were defined in a co-culture of MKN-45 as a target cell (T). The ratios of 1:1, 1:3, and 1:6 were studied. LDH assay was performed after 24 h of incubation as a necrosis marker in a cell culture medium (Chan, Moriwaki et al., 2013).

All animals were housed in individually ventilated cages (IVC) for animal experiments. The center maintains standardized housing conditions and hygiene management according to the guidelines of the ethics committee. The animals are preserved under special conditions, e.g., a 12/12 h light-dark cycle, a humidity of 65%, and a temperature of 25°C. To assess the antitumor effect of NK cells in combination with Nivolumab (Bristol Myers Squibb, USA) in vivo, we used 6–8-week-old female NOD. Cg-Prkdcscid IL2rgtm1Sug (NOG) mice were obtained from the animal facility of the Digestive Disease Research Institute of Tehran University of Medical Sciences, which subcutaneously inoculated by 5×10⁶ MKN-45 cells. Prior to euthanasia, animals are anesthetized with ketamine and xylazine and euthanized with CO₂. Tumor sizes and body weights were measured twice a week. Animals with a tumor volume ≥2000 mm³, real bodyweight loss ≥20%, or BC = 1 (body condition) were humanely terminated. All animal experiments were authorized by the institutional ethics committee of the Tehran University of Medical Science (IR.TUMS.MEDICINE.REC.1399.644).

Tumor median size of 100–200 mm³ (day 0) was selected for dividing the animals randomly into four experimental groups, including IL-2-activated NK cells (NK cells), IL-2-activated NK cells along with anti-PD-1 antibody (Nivolumab) (NK + anti-PD-1), and IL-2-activated NK cells ex vivo pretreated with 20 μg/ml anti-PD-1 (NK-anti-PD-1 pretreated). Each mouse received 10×10⁶ NK cells twice at 2-week intervals via the IV route (Figure 1). The dosage of anti-PD-1 (Nivolumab) was translated from human into murine setting based on body surface area using the following formula (Nair and Jacob 2016): Murine equivalent dosage ( unit / kg ) =  pediatric dosage ( unit / kg ) × H u m a n   K m M o u s e   K m . The Km constant was 37 for adult humans 37 and 3 for mice (Reagan-Shaw et al., 2008). The NK + anti-PD-1 group was treated with 37 mg/kg anti-PD-1 antibody twice at 2-week intervals via the IV route. Control is the fourth group, which received phosphate-buffered saline (PBS).

Tumor volume was measured by a caliper and calculated by the following formula: tumor volume = 1/2 (length × width 2). Relative tumor volume (RTV) was evaluated for morphometric growth kinetics of the tumors by dividing tumor volume on a measured day to tumor volume on day 0. Antitumor activity was presented by tumor growth inhibition (TGI) percentage using a formula by Tsukihara et al.: TGI (%) = [1 − (RTV of the treated group)/(RTV of the control group)] × 100 (%) (Tsukihara, Nakagawa et al., 2015).

NK cells characterization by PE anti-human CD56⁺ and FITC anti-human CD3⁻ markers was quantitatively evaluated; the infiltration of NK cells at the tumor site was performed using flow cytometry (FACSCalibur Becton Dickinson, United States). NKG2D and CD69 were evaluated before and after treatment of IL-2-stimulated NK cells by the anti-PD-1 antibody for further characterization. The whole tumor was collected and digested mechanically and enzymatically, and staining was done according to the manufacturer's instructions; then, flow cytometry was performed (Feng, Peng et al., 2009). Data were analyzed using the FlowJo 7 software (Tree Star, Inc.). Experiments were carried out in triplicate and quantified by GraphPad Prism 9 (GraphPad Software Inc., CA, United States).

The histopathologic response of tumors to the interventions was assessed based on the residual tumor (R) (Edge, Byrd et al., 2010) (Table 1). Hematoxylin and Eosin (H and E) stain slides were prepared to stratify the histopathologic responses. Proliferative activity changes assessed mitotic count on H and E slides. As previously described by Meuten and colleagues, mitotic cells are counted in areas and reported as an average over ten consecutive high power fields (HPF) (Meuten, Moore et al., 2016).

Immunohistochemistry (IHC) against human CD56 (Biolegend, United States) was performed as described previously (Kim et al., 2016) for NK cell infiltration evaluation. Furthermore, the active protein of caspase 3 for apoptosis induction evaluation was detected using a rabbit polyclonal antibody (Biolegend, United States). IHC was assessed based on the Allred score; this scoring combines the percentage of immunoreactive cells and the intensity score (Parvin et al., 2019) (Table 2). Following the euthanasia of the mice, tumor tissues were collected and processed for paraffin embedding. Briefly, paraffin-embedded tumors were cut by microtome to a thickness of 4 μm. After deparaffinization and hydration, the tumor tissues were incubated by primary and secondary antibodies (Adan, Alizada et al., 2017).

All results are reported as the mean ± SEM and the data were analyzed by GraphPad Prism 9 software package (GraphPad Software, Inc., San Diego, United States). A p value of <0.05 represented statistical significance.

NK cells isolated from peripheral blood were characterized by the presence of CD56 and the absence of CD3 as NK cell surface biomarkers. The purity was 93.4%, which included the CD3⁻ CD56⁺ cells based on flow cytometric analysis. Freshly isolated CD3- and CD56+ cells were cultured in the presence of IL-2 and feeder cells and expanded for 21 days. Expanded NK cells exhibited clonal growth and round shape morphology under cytokine-/feeder-enriched (IL-2/EBV-LCL cells) medium; finally, after 21 days, we reached 370 million cells with more than 95% viability, which were evaluated by trypan blue before injection (Figures 2A,B).

Further investigation of NK + anti-PD-1 immunophenotyping NKG2D and CD69 was performed using flow cytometry. NK NKG2D⁺ and CD69⁺ cells increased 10 and 8 percent after treatment with IL-2-stimulated NK cells with anti-PD-1 antibody (p < 0.0007 and p < 0.0085, respectively). Moreover, quantification of mean fluorescence intensity (MFI) showed that anti-PD-1 treatment improved CD69 expression (p < 0.002), but no significant changes were observed in NKG2D marker after treatment with IL-2-stimulated NK cells with anti-PD-1 (Figures 2C,D,E).

To evaluate the effect of a combined strategy of IL-2-activated NK cells with anti-PD-1 antibody in vitro, MKN-45 cells were co-cultured with activated NK cells at three specific E: T ratios (1:1, 3:1, and 6:1). After a 24 h incubation, the NK-cell-mediated cytotoxicity against MKN-45 was detected by LDH assay. In the non-treated group, NK cells showed 11, 16, and 25 percent cytotoxicity at the following E: T ratios: 1:1, 3:1, and 6:1, respectively. In the anti-PD-1-treated group, the lysis percentages were 13, 19, and 30. The most prominent antitumor cytotoxicity of activated NK cells was achieved when PD-1 was inhibited with an anti-PD-1 antibody (p < 0.006) (Figure 3).

In order to explore the effect of adaptive NK cell therapy combined with Nivolumab in tumor growth in vivo, we further investigated the impact of interventions on the morphometric properties in the subcutaneous transplantation mouse model of gastric cancer using MKN-45 cell line.

The use of an Anti-PD-1 antibody confirms the hypothesis that PD-1/PDL-1 pathway inhibition can modulate the therapeutic cytotoxicity effect of NK cells. Ten days after cell line inoculation, tumors reached the desired size of 100–200 mm³ for interventions. The sham-treated mice showed the most rapid tumor growth. Two IV injections of ex vivo IL-2-activated NK cells combined with an anti-PD-1 antibody caused a significant tumor growth delay. The morphometric growth curves of tumors are shown in Figure 4A. Optimal percent of TGI was observed 28 days after the beginning of treatment. The highest tumor growth inhibition was considered in the NK-anti-PD-1 pretreated group, with a median TGI of 38% (31–41). Also, the NK + anti-PD-1 therapy group showed an inhibitory effect with a median TGI of 18% (12–19); NK cells as monotherapy did not show an inhibitory effect on the gastric cancer model with a median TGI of -0.5%. NK-anti-PD-1 pretreated group significantly inhibited tumor growth more than NK + anti-PD-1 and NK cells (p < 0.01 and p < 0.001, respectively). Furthermore, NK + anti-PD-1 induces tumor inhibition compared to NK cells (p < 0.03). The best therapeutic outcome was observed after modulating NK cells by anti-PD-1 antibody (Figure 4B).

PD-1 blockade was used to improve response to the immune-cell-based therapy approach. After implantation of 5×10⁶ MKN-45 cells, the mice bearing tumor was intravenously injected with 10×10⁶ NK cells, NK-anti-PD-1 pretreated, NK + anti-PD-1, and PBS (control).

NK-anti-PD-1 pretreated group based on residual tumor showed that in more than 30% of tumors, the fibrosis areas and apoptotic fragment were increased, R2 histologic response (p < 0.0001), in response to intervention. Furthermore, the NK + anti-PD-1 group showed a low response rate, R3 histologic response (p < 0.02). NK cells showed a degree of histopathologic response, which not statistically significant (Figures 5A,B).

Mitotic count was investigated to evaluate the effect of interventions on tumor proliferation intensity. NK cells anti-PD-1 pretreated showed the highest anti-proliferative effect; the mitotic count in this group was significantly lower than that in the control group (12 [8–15] vs. 21.5 [19–35] (p < 0.002). Furthermore, the mitotic count of the NK + anti-PD-1 group was significantly decreased compared to the control 14 [9–21] (p < 0.03), but NK cells did not show a difference in mitotic count (Figure 5C).

The cytotoxicity mechanism of different treated groups via apoptosis was evaluated by caspase 3 labeling. Our findings based on Allred score show that NK cells-anti-PD-1 pretreated and NK-Anti-PD-1 groups have intermediate level of caspase 3, while NK cell alone was weakly positive for caspase 3 (Figure 6).

In order to verify the NK cell permeation, human CD56⁺ cells were evaluated. NK-anti-PD-1 pretreated cells were intermediate positive and the other treated groups that received NK cells were weakly positive for CD56 biomarker based on Allred score (Figure 7A). The infiltration percentage of the NK cells, NK + anti-PD-1, and NK-anti-PD-1 pretreated cells groups were 3, 5, and 7%, respectively. Thus, this result indicated that NK-anti-PD-1 pretreated group showed the highest infiltration (p < 0.008) (Figure 7B).