ENKTL tumors (n = 43) were obtained from the Samsung Medical Center. Baseline patient characteristics collected for analysis included age, sex, Eastern Cooperative Oncology Group (ECOG) performance status, Ann Arbor stage, involvement of the nasal cavity or nasopharynx, and regional or distant lymph nodes. Nasal and non-nasal types were defined based on involvement of the nasal area47. A prognostic index for ENKTL was assessed as low-risk, intermediate-risk, or high-risk groups based on the prognostic index for natural killer cell lymphoma (PINK)48. PINK-Epstein-Barr virus (PINK-E) was obtained for patients with data for peripheral blood EBV DNA status. Any detectable concentration of EBV DNA was defined as positive. Of the 43 total patients, 30 had information on immune subtyping to classify ENKTL patients into four tumor immune microenvironment subgroups of immune tolerant (IT), immune evasion-A (IE-A), immune evasion-B (IE-B), and immune silenced (IS)19.
All patients provided written informed consent for the use of archival tissues with retrospective clinical data. This study was performed in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Samsung Medical Center.
All available hematoxylin and eosin (H&E)-stained slide from archival formalin-fixed, paraffin-embedded (FFPE) tissues were reviewed by two pathologists (H.B. and Y.H.K.). Total RNA was extracted from 2 to 4 sections of 4-μm-thick sections from whole FFPE tumor tissues using the High Pure RNA Paraffin kit (Roche Diagnostic, Mannheim, Germany). The RNA was quantitated using UV spectroscopy (Nanodrop Technology, Wilmington, DE, USA).
NanoStringⓇ nCounter assay using a probe for 137 genes
NanoString-based multigene assay was performed according to the published literature49. Briefly, 200 ng total RNA was used to determine gene expression levels utilizing NanoString technology (NanoString Technologies, Seattle, WA, USA). An nCounter CodeSet (NanoString Technologies) containing a biotinylated capture probe for 133 target genes and 4 housekeeping genes (ACTB, B2M, G6PD, and GAPD; Supplementary Table 3) was used for gene expression analyses. The data were normalized to the mean expression levels of internal reference genes with a cutoff value of 20. Standard quality control was performed with nSolver™ Analysis Software (NanoString Technologies) with flagging of any sample with a total of the positive spike-in controls outside of 0.3–3 times the geometric mean of the total positive spike-in for that cartridge. Two samples in these experiments were flagged. The probe counts were then normalized using the geometric mean of the 4 housekeeping genes and log2 transformed for further analysis.
SNK6 was kindly provided by Dr. Y. K. Jeon (Seoul National University Hospital, Seoul, Korea) and NK92MI were purchased from American Type Culture Collection (Rockville, MD, USA). SNK6 was cultured in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 500 U/mL of interleukin-2. NK92MI was cultured in MEM-α medium supplemented with 20% heat-inactivated FBS. Penicillin and streptomycin (Gibco-BRL, Grand Island, NY, USA) were added to the media and cells were incubated in a humidified 5% CO2 atmosphere. Small interfering RNAs (siRNAs) were purchased from Bioneer (Daejeon, South Korea). Cells were transiently transfected using Lipofectamine RNAiMAX (Invitrogen, Carlsbad, CA, USA). Doxorubicin was purchased from Apexbio (Boston, MA, USA).
Quantitative reverse transcription polymerase chain reaction (qRT-PCR)
Total RNA was prepared using a Qiagen RNA extraction kit (Qiagen, Valencia, CA, USA), according to the manufacturer’s instructions. For reverse transcription, 1 μg of RNA was treated with RNase-free DNase, and cDNA obtained using an Omniscript RT kit (Qiagen). The generated cDNA was amplified using primers specific for EGR-1, CD59, CXCR7, GAS1, and RAMP3 (Supplementary Table 4). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was amplified as the control.
Apoptosis was detected using an Annexin V-fluorescein isothiocyanate (FITC) Kit (BD Biosciences, San Jose, CA, USA) and a BD FACS ARIA III flow cytometry system (BD Biosciences). Cells were exposed to doxorubicin or gamma irradiation, then harvested and processed according to the manufacturer’s instructions.
A two-sample t-test was used to determine differential expression of genes between groups. The t-test P-value of each gene was calculated based on 10,000 permutations and the adjusted P-value was calculated using the single-step procedure to control the family-wise error rate (FWER)50. The Kaplan-Meier method was used to estimate OS rates and the log-rank test was used to compare survival distributions between the groups. The multivariate Cox regression analysis was used to identify risk factors associated with significant genes. The optimal cutoff was selected as the point with the most significant log-rank P-value for all possible cutoff points. Statistical analysis was performed using R 3.0.2 (Vienna, Austria; http://www.R-project.org/) and SAS 9.4 (SAS Institute, Cary, NC, USA).
When reporting this study, we adhered to the guidelines of an important methodological paper from 2005 entitled “Reporting recommendations for tumor marker prognostic studies (REMARK guidelines)”51,52.