All cell lines were purchased at ATCC (ATCC/LCG, Wesel, Germany) except PC9 (PHE culture collection, Salisbury, UK) and Nuli-1 (donated from Professor Christian Holm, Aarhus University, Denmark). Cells were grown in RPMI (Sigma-Aldrich, Cat#: R8758) or DMEM (Sigma-Aldrich, Cat#: D6429) according to supplied instructions supplemented with 10% fetal bovine serum (Sigma-Aldrich, Cat#: F9665), 1% L-Glutamine (Thermo Fisher Scientific, Cat#: 25030024) and 1% Penicillin-streptomycin (Thermo Fisher Scientific, Cat#: 15140122). Nuli-1 cells were grown in culture flasks precoated with collagen (Sigma-Aldrich, Cat#: C7521) in Bronchial Epithelial Cell Growth Basal Medium (Lonza, Cat#: CC-3171) with supplements and growth factors (Lonza, Cat#: CC-4175). The cells were grown at 37°C and 5% CO₂.

For all experiments including transfection, cells were seeded one day prior to transfection in a suitable plate size (Nunc 12-well, Nunc 24-well or Nunc 96-well plates). Lipofectamine® 2000 (Invitrogen, Cat#: 11668019) was used for transfection with 2μg/mL double stranded herring testis DNA (HT-DNA) (Sigma-Aldrich, Cat#: D6898-250) or 40ng/ml polyinosinic-polycytidylic acid (Poly I:C)-LMW (Invivogen, Cat#: tlrl-picw) according to the manufacturer’s instructions.

For all experiments using cisplatin and doxorubicin, an approximated IC₅₀ dose was used. Due to stability of the drugs, the exact concentration of the drugs corresponding to an IC₅₀ dose varied throughout the course of the experiments. For cisplatin, the following dose ranges have been given to the individual cell lines; H358: 3-16μM, H1650: 3.34-8.5μM, H2228: 0.6-7.3μM, and H596: 1.6-8.5μM. For doxorubicin, the following dose ranges have been given to the individual cell lines; H358: 0.5-0.8μM, H1650: 0.5-1.1μM, H2228: 0.5-0.95μM, and H596: 0.5-1.1μM.

IC₅₀ values were determined by seeding 5000-15000 cells in a Nunc 96-well plate and 24 hours later treating the cells for 48 hours with a dilution range of either cisplatin or doxorubicin. Viability as percentage of an untreated control was determined using CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega, Cat#: G3580) according to manufacturer’s instructions. A regression analysis was performed to estimate the IC₅₀ value.

In Supplementary Figures 1A, B , type I IFN production was measured using a biofunctional Human HEK-Blue IFN-α/β reporter cell assay (Invivogen). Fifty microliters of standard (IFN-α standard curve starting at 1000U/mL (PBL Assay Science, Cat#: 11100-1)) or supernatants from the stimulated cells were added to the HEK-Blue IFN-α/β reporter cells overnight. Supernatant (20 μl) from HEK-Blue IFN-α/β reporter cells were then mixed with 180 μL QUANTI-blue (Invivogen, Cat#: rep-qb2) and the optical density at 620 nm was determined using a microplate reader to determine final concentration as U/mL.

Detection of CCL5/RANTES, IFNβ, IFNλ, and IL-6 in supernatants was caried out using Human DuoSet ELISA (R&D systems, Cat#: DY278, DY814, DY7246, DY206) according to the manufacturer’s instructions. The optical density was determined using a microplate reader set to 450 nm and 570 nm. The readings at 570 nm were subtracted from the 450 nm-reading to correct for optical imperfections in the plate.

IL-6 and CXCL10 in Supplementary Figures 2C–G were detected using Multiplex ELISA from Meso Scale Discovery (MSD) U-plex platform according to manufacturer’s protocol (Meso Scale Diagnostics, Cat#: K15067L-2).

Samples were harvested in RIPA Lysis and Extraction Buffer (Thermo Scientific, Cat#: 89900) supplemented with protease and phosphatase Inhibitor (Millipore Sigma, Cat#: 11873580001 and Thermo Fisher Scientific, Cat#: A32961) and 10mM NaF (VWR, Cat#: J60251.AE). Before loading, the samples were mixed 1:1 with Laemmli buffer (Sigma Aldrich, Cat#: 38733) and boiled at 95°C for 5 min. The samples were loaded on a 10% Criterion™ TGX™ Precast Midi Protein Gel, 26 well (Bio-Rad Cat#: 567-1035) and run in MOPS Running Buffer (Thermo Fisher Scientific, Cat#: NP0001). The gel was blotted onto a Turbo Transfer Midi PVDF membrane (Bio-Rad Cat#: 170-4157) and the membrane was blocked with 5% skimmed milk (Sigma-Aldrich, Cat#: 70166). The membrane incubated overnight with primary antibody with rotation at 4°C, and hereafter incubated with secondary antibody for 1hour with rotation at room temperature before development with Clarity Western ECL Substrate (Bio-Rad, Cat#:1705060) or SuperSignal West Femto Maximum (Thermo Fisher Scientific, Cat#: 34095) using the Azure Biosystems imaging system 300. Primary antibodies were all used in ratio 1:1000 with 5% Bovine Serum Albumin (Roche, Cat#: 10735986001) except Vinculin, which was diluted 1:10000. Primary antibodies from Cell Signaling Technology: anti-TBK1 Cat#: 3313S (84 kDa), anti-STING Cat#: 13647S (33-42 kDa), anti-cGAS Cat#: 15102S (62 kDa), anti-IRF3 Cat#: 4302S (55 kDa), anti-pTBK1 Cat#: 5483S (84 kDa), anti-pSTING Cat#: 19781S (33-42 kDa), anti-STAT1 Cat#: 14994S (90 kDa), anti-pSTAT1, Cat#: 7649S (90 kDa). From Sigma Life Sciences: anti-Vinculin Cat#: V9131 (116 kDa) and from Abcam: anti-pIRF3 Cat#: 76493 (55 kDa). Secondary antibodies from Jackson ImmunoResearch: Peroxidase-AffiniPure F(ab’)2 Fragment Donkey Anti-Rabbit IgG Cat#: 711-036-152, Peroxidase-AffiniPure F(ab’)2 Fragment Donkey Anti-Mouse IgG Cat#: 715-036-150.

Genetically modified STING knock-out (KO) cell lines were generated using chemically modified guide RNA’s (sgRNAs) in combination with the nuclease, spCas9. Double strand breaks generated by spCas9 resulted in the creation of indels and ultimately knocking down the gene expression. sgRNAs were designed using the bioinformatic platforms CRISPick and CRISPOR and purchased from Synthego with a chemical 2’-O-methyl-3’phosphorothioate-modification on three terminal nucleotides in both ends. We designed two individual sgRNAs targeting approximately 50bp apart to increase the chances of a complete knockout of STING. Initially, ribonicleoprotein (RNP) complexes were formed by incubating 6μg spCas9 protein (Alt-R S.p. Cas9 Nuclease V3, Integrated DNA Technologies) with 3.2μg of each individual sgRNA for 15min. at room temperature. Hereafter, 100,000 cells were mixed with the RNPs and electroporated on the 4D-nucleofector (Lonza) in Nucleovette strips (Lonza) using the program CM138-P3. Cells were hereafter cultured under the same conditions as wildtype. sgRNA sequences provided in Supplementary Table 2 .

RNA was purified using RNeasy mini kit (Qiagen, Cat#: 74106) or NucleoSpin RNA kit (Macherey-nagel, Cat#: 740955) and eluted in 30-40μL RNase-free water. 250ng-1μg of RNA was used for cDNA synthesis using the iScript cDNA synthesis kit (Bio-Rad Cat#: 1708891BUN) on an Arktik thermal cycler (Thermo scientific) with program: 5’25˚C; 20’46˚C; 1’95˚C; 4˚C.

To determine expression levels of the human IFNB1, IFNL1, IFNL2, IFNLR1 YWHAZ and EIF2B2 on lung cancer cell lines, gene specific primers were designed using Primer3 and purchased from Eurofins ( Supplementary Table 3 ). Samples were analyzed in a final volume of 10μL, containing 5μL KAPA SYBR Fast Master Mix (Sigma-Aldrich, Cat#: KK4611), 2μM forward primer, 2μM reverse primer, nuclease-free water and 10ng of cDNA. Analysis was performed on a Lightcycler 480 platform with program: 3’95˚C; 40x (10”95˚C; 20”60˚C, 1”72˚C). Ct values were extracted using the Lightcycler Software. For obtaining gene expression values X0 was calculated as 2^-Ct and normalized to the X0 value of a reference gene. For each sample, two-three technical replicates were included and averaged during gene expression calculation.

Samples for RNA sequencing are generated in one experiment performed in duplicates. Cells were either left untreated or transfected with 2μg/mL of HT-DNA as described in the paragraph about transfection above. RNA was purified using RNeasy mini kit (Qiagen, Cat#: 74106) or NucleoSpin RNA kit (Macherey-nagel, Cat#: 740955) and eluted in 30-40μL RNase-free water.

RNA was sequenced either at BGI Copenhagen or Novogene Europe. A strand-specific and polyA-selected mRNA library was prepared, and the samples were sequenced at a depth of either 30 million reads per sample using paired-end reads of 100bp on a DNBSEQ platform (BGISEQ) or 20 million reads per sample using paired-end reads of 150bp on a Novaseq X plus platform.

The output fastq files were analyzed with R version 4.4.1 or the CLC Genomics Workbench 12 (QIAGEN) software. Using R or the RNA-seq module of the CLC Genomics Workbench 12, all reads were trimmed and mapped to the Genome Reference Consortium Human Build 38 (GRCh38/hg38). Expression values of each gene were normalized to transcripts per million (TPM).

TPM values from RNA sequencing were analyzed in the R software version 4.4.1. Initially all genes where all the samples had a TPM value < 5 were excluded. Hereafter a value of 1 was added to all TPM values to avoid zeros. A log2 fold change between the untreated and the HT-DNA transfected sample was calculated for each gene and each cell line. The average log2 fold change for each gene across the six cell lines was then calculated to rank the genes according to log2 fold change. The 50 genes with the highest log2 fold change were then included in a heatmap displaying the Z-score for each cell line, untreated (UT) and HT-DNA transfected (DNA).

A comprehensive dataset of lung cancer single cell data from 316 patients, already processed and integrated (42), was downloaded from the Cell x Gene platform. We filtered the data on two parameters: 1) The samples should be either from lung cancer patients or healthy controls. 2) The samples were of lung tissue. The cell types were pre-annotated, and we simplified the cell type annotation as shown in Supplementary Table 4 . The expression of IFNLR1 and IL10RB was visualized for both the healthy controls and cancer samples. For the visualization, the cells without expression for the two genes were removed. The difference in expression between the groups were evaluated by a Wilcoxon Rank Sum test calculated using the R package ggpubr. (Kassambara A (2023) ggpubr: ‘ggplot2’ Based Publication Ready Plots (R package version 0.6.0). https://CRAN.R-project.org/package=ggpubr).

Transcriptional activation of IFNLR1 was carried out using chemically modified sgRNAs in combination with a mutant spCas9 nuclease with a defective nucleolytic activity (dCas9). Moreover, the dCas9 was fused to the three transcription activators p65, Rta, and VP64 (dCas9-VPR). sgRNAs were designed using the bioinformatic platforms CRISPick and purchased from Synthego with a chemical 2’-O-methyl-3’phosphorothioate-modification on three terminal nucleotides in both ends.

In total eight different sgRNAs targeting the promoter region of IFNLR1 were tested both alone and some in combination to determine which single sgRNA or combination of sgRNAs would result in the best transcriptional activation of IFNLR1. The dCas9-VPR mRNA was in vitro transcribed from a plasmid provided by Rasmus O. Bak (43).

For experiments including cells with transcriptionally activated IFNLR1, 1μg or each sgRNA was mixed with 0.5μg dCas9-VPR mRNA and added to 500.000 cells. A mock control with dCas9-VPR mRNA but no sgRNA is included in all experiments with CRISPRa. To deliver the CRISPRa components, cells were then electroporated on the 4D-nucleofector (Lonza) in Nucleovette strips (Lonza) using the program CM138-P3. sgRNA sequences provided in Supplementary Table 5 .

For determining viability 5000 cells were seeded per well in a NUNC 96-well plate. 24 hours after seeding, the cells were treated with either IFNλ or IFNβ for 24 hours. Viability as percentage of an untreated control was determined using CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega, Cat#: G3580) according to manufacturer’s instructions.

Apoptosis resulting from caspase 3/7 activation after 48 hours treatment with approximated IC₅₀ doses of cisplatin (Sigma-Aldrich, Cat#: 232120) and doxorubicin (Sigma-Aldrich, Cat#: D1515) was detected by flow cytometry. Cells were washed three times in PBS, trypsinized and resuspended in cell culture medium. Hereafter, cells were pelleted by centrifugation at 400xg for 5 minutes, resuspended in PBS and counted. Viability was determined from these cell counts. In each individual experiment, an equal number of cells were analyzed for each condition in a V-bottom 96-well plate (Sarstedt). Cells were resuspended in 95μL of PBS supplemented with 0.5% Bovine Serum Albumin (Roche, Cat#: 10735986001), 0.09% sodium azide (Sigma-Aldrich, Cat#: S2002), and 1mM EDTA (Invitrogen, Cat#: 15575020). To stain the cells for Caspase 3/7 activation, 5μL of Caspase-3/7 Reagent (Invitrogen, Cat#: R37111) were added to 95μL cell suspension and incubated in the dark for 30 minutes. Right before analyzing the samples, 2.5μL of 50μg/mL Propidium iodide Solution (Sigma-Aldrich, Cat#: P4864) were added to the samples. Cells were analyzed by flow cytometry measuring fluorescence emission at 530nm and 615nm using a NovoCyte Quanteon 4025 flow cytometer equipped with four lasers (405 nm, 488 nm, 561 nm and 637 nm) and 25 fluorescence detectors (Agilent, Santa Clara, CA). Data analysis was performed using FlowJo version 10.10.0. The gating strategy can be found in Supplementary Figures 4 , 5 .

The sample size and statistical methods used for each experiment have been provided in the respective figure legends. Data was analyzed with GraphPad Prism 10.2.3 (GraphPad Software) and R version 4.4.1. In cell experiments, comparison between groups was performed using Two-way ANOVA analysis using Šídák’s multiple comparisons test or multiple T tests with Šídák-Holm’s correction for multiple testing. For single cell RNA sequencing data, comparison of groups is done using Wilcoxon Rank Sum test. * = P <.05, ** = P <.01, *** = P value <.001, **** = P <.0001.

Figures were prepared using Biorender.com and PowerPoint.

As STING protein expression is a prerequisite for pathway activity ( Figure 1A ), we conducted an initial screening of STING expression in 11 different NSCLC cell lines. This revealed that approximately half of the cell lines did not express STING while the cell lines HCC827, H358, H1650, H596, H1975, and H2228 did ( Figure 1B ). We next evaluated the functionality of STING in each of the cell lines by measuring type I IFN and IL-6 in response to exogenous DNA stimulation, in form of lipofectamine delivered herring-testis (HT)-DNA. As control, we included the TLR3 agonist, Poly I:C, which induces secretion of the measured cytokines by relying on the same transcription factors ( Supplementary Figures 1A–D ). We were only able to detect type I IFN from four of the six STING-expressing cell lines (H1650, H596, H358 and to a lesser extent in H1975) in response to HT-DNA ( Supplementary Figure 1A ). These and additional two cell lines did also produce type I IFN in response to Poly I:C ( Supplementary Figure 1B ). All STING-expressing cell lines produced IL-6 (to various degrees) in response to HT-DNA stimulation, but interestingly, so did two of the non-STING expressing cell lines (PC9 and H1568; Supplementary Figure 1C ). All, with exception of two cell lines (A427 and A549), did secret IL-6 in response to Poly I:C ( Supplementary Figure 1D ). These results suggested that type I IFN is a poor marker for STING pathway activation in NSCLC.

We next investigated the transcriptomic profile induced by HT-DNA in the six STING-expressing NSCLC cell lines by RNA sequencing. A waterfall plot of average log2 fold change across all cell lines showed that DNA stimulation both decreased and increased gene expression although the increase led to more genes with a higher log2 fold change ( Supplementary Figure 1E ). In fact, 87 genes had an average log2 fold change above 2 whereas 0 genes were found to have an average log2 fold change below -2 ( Supplementary Table 1 ). An inspection of the 50 most upregulated genes across the cell lines showed a classical IFN-signature with expression of ISGs like IFIT2, IFIT3, RSAD2, IFIT1, OASL, ISG15, PLAUR and KLF4. Interestingly, the average upregulation of the less studied type III IFNs, IFNL1 and IFNL2, were both in the top 10 of upregulated genes together with IFNB1. It was also apparent that interferon expression in general was very cell type dependent with H2228 not upregulating IFNB1 at all and H1975, HCC827, and H358 only to a lesser extent in response to HT-DNA stimulation. In terms of IFNL1, it was found to be upregulated to a higher degree across all cell lines ( Figure 1C ). This observation was confirmed when we compared expressions of IFNL1, IFNL2 and IFNB1 across eight NSCLC cell lines after HT-DNA stimulation. Here IFNL1 was relatively higher expressed compared to IFNB1, with exception of H596. For the two STING-negative cell lines PC9 and H1568 both type I and III IFNs were undetectable ( Figure 1D ).

Next, we conducted a deeper characterization of four STING expressing cancer cell lines and compared them to the immortalized healthy epithelial cell line Nuli-1. All cancer cell lines as well as Nuli-1 showed phosphorylation of STING, TBK1, and IRF3 after stimulation with HT-DNA, indicating an active signaling pathway ( Figures 1E, F ). However, H596, H1650 and Nuli-1 were the only cell lines producing IFNβ in response to HT-DNA stimulation ( Figures 1G–I ). Interestingly, IFNβ could be detected upon Poly I:C stimulation in H358 but not at all in H2228 ( Supplementary Figure 2A ). All four cancer cell lines had detectable IFNλ levels upon HT-DNA as well as Poly I:C stimulation ( Figure 1H , Supplementary Figure 2B ). Additionally, all cell lines also produced IL-6 and CCL5 in response to HT-DNA, whereas H1650 and H2228 were not able to produce CXCL10 ( Supplementary Figures 2C–G ).

Taken together, these results suggests that IFNλ could be a more relevant marker for an active STING pathway in NSCLC compared to type I IFN and CXCL10.

Since we found the different cancer cell lines to generally secrete more type III IFNs than type I IFN, we wanted to determine if this correlated to alternative gene expression kinetics. Thus, we stimulated each cancer cell line with HT-DNA and measured expression of IFNL1, IFNL2, and IFNB1 after 2, 6, 10 and 20 hours ( Figures 2A–E ). In accordance with our earlier findings, we observed that IFNL1 was relatively higher expressed than IFNB1 in three of the four NSCLC cell lines and followed an expression kinetic pattern like IFNB1 ( Figures 2A–D ). The cell line H596 and Nuli-1 did stand out having higher IFNB1 expression ( Figures 2B, E ). These observations were mirrored in the protein secretion pattern ( Figures 2F–J ).

Sensing of endogenous DNA in cancer cells through the cGAS-STING pathway facilitated by DNA-damaging agents has been proposed to initiate anti-tumoral responses ( Figure 3A ). From our initial characterization of the four NSCLC cell lines, we observed a diverse pattern of cGAS protein levels with H358 expressing the most and H1650 the least cGAS ( Figure 1E ).

To determine how IFNL1 and IFNB1 expression was regulated upon DNA damage-inducing agents, each cell line was treated with doxorubicin or cisplatin for 48 hours. Interestingly, IFNL1 expression was significantly upregulated by both doxorubicin and cisplatin in three of the cancer cell lines and did not seem to correlate with cGAS protein expression ( Figures 3B–E ). Once again H596 stood out being the only cell line demonstrating significant upregulation of only IFNB1 and not IFNL1 expression ( Figure 3C ).

To confirm that responses were truly STING-dependent, we next used CRISPR-Cas9 technology to generate a STING knockout (KO) variant of each NSCLC cell line. The level of STING depletion was confirmed by western blot ( Figure 4A ). Next, a control KO (mock) and STING KO cells were treated with doxorubicin ( Figures 4B–E ) or cisplatin ( Figures 4F–I ) for 48 hours followed by expression analysis of IFNB1 and IFNL1. Overall, these results confirmed that the DNA damage-inducing agents triggered IFNB1 and IFNL1 expression in a STING-dependent manner.

A major difference between type I and type III IFNs is how they modulate a local environment through autocrine and paracrine receptor interactions. We used the human protein atlas database to examine the expression pattern of IFN receptor subunits. The type I IFN receptor subunits, IFNAR1 and IFNAR2, as well as type III IFN receptor subunit IL-10Rβ are ubiquitously expressed in our body whereas the type III IFN receptor subunit IFNLR1 is expressed at low levels and more selectively specific organs ( Figure 5A ). Furthermore, exploration of numerous cell types demonstrated that IFNLR1 expression is very sparce and clearly cell type specific ( Figure 5B ). Next, using a publicly available single cell RNA sequencing dataset from lung cancer tumors (42), we found IFNLR1 expression to be significantly decreased in malignant epithelial cells compared to normal epithelial cells from the same tumor ( Figure 5C ). Furthermore, we could demonstrate that IFNLR1 expression was lower in our studied NSCLC cell lines compared to Nuli-1 ( Figure 5D ), suggesting that IFNLR1 is selectively downregulated in NSCLC both in patient tumors, and in in vitro cell line models. Next, we wanted to explore the ability of IFNλ and IFNβ to induce an IFN-mediated response in our NSCLC cell lines. Knowing that IFNλ and IFNβ activates the same downstream signaling pathway after receptor engagement involving phosphorylation of STAT1 we stimulated two of our NSCLC cell lines with recombinant IFNλ and IFNβ to evaluate STAT1 phosphorylation. Here, we only identified phosphorylation of STAT1 in response to IFNβ ( Figure 5E ).

The selective downregulation of IFNLR1 expression in cancer cells could be seen as an immune escape mechanism evolved during cancer development which allows cancer cells to secrete IFNλ to support immunomodulation of the tumor microenvironment while avoiding being affected themselves.

Gene regulation can be achieved by various epigenetic drugs and is already used in various cancer settings (44). However, most drugs are often not specific to a single gene of interest but affects multiple genes within a single cell. Thus, to investigate the specific role of IFNLR1 expression and pathway activation in NSCLC, we established a CRISPR-mediated transcriptional activation approach (CRISPRa) ( Figure 6A ) targeting the promotor region of IFNLR1. We combined mRNA of dead Cas9-VPR with a total of eight different guide RNAs (sgRNAs) which we tested individually and in combination to identify the best sgRNA(s) for transcriptional activation following electroporation of the RNA molecules ( Supplementary Figure 3A ). Ultimately a combination of two gRNAs resulted in a clear upregulation of IFNLR1 and was associated with robust receptor signaling measured by phosphorylation of STAT1 (pSTAT1) ( Supplementary Figure 3B ). Next, we evaluated the kinetics of IFNLR1 mRNA expression and pSTAT1 signaling following CRISPRa in H1650 cells. The gene expression of IFNLR1 peaked one day after electroporation and was barely detectable after four days ( Figure 6B ). Regarding receptor activation in the cells, we observed pSTAT1 expressed up to three days after stimulation with IFNλ ( Figure 6C ).

It is well known that interferon signaling can affect cancer cell viability negatively (37–39, 45–47). To test this in our settings, we first treated cells with recombinant IFNβ as we already knew that the cancer cell lines were sensitive to IFNβ signaling ( Figure 5E ). This resulted in a minimal decrease in cell viability of 10-20% compared to untreated cells ( Figures 6D, F ) but not in a dose-dependent manner. When treating with recombinant IFNλ, both cancer cell lines were unaffected by the treatment but upon CRISPRa of IFNLR1 we observed a significant drop in cell viability of 20-40% ( Figures 6E, G ). These results encouraged us to investigate the potential of transcriptionally upregulating IFNLR1 as a way of sensitizing NSCLC cells to endogenous IFNλ secretion.

Cisplatin induces inter- and intrastrand cross-links (48), and moreover contributes to cellular disfigurement from oxidative stress and DNA damage. We also found that cisplatin treatment of our NSCLC cell lines induced measurable IFNλ levels ( Figures 3B–E ). Thus, to evaluate if chemotherapy-induced type III IFN production was enough to trigger IFNLR1-mediated apoptosis, we finally performed transcriptional upregulation of IFNLR1 (IFNLR1+) in combination with cisplatin treatment for 48 hours. In our two NSCLC cell lines, we observed that cisplatin treatment resulted in a decreased viability measured by cell count and this seemed to be slightly increased in IFNLR1+ cells ( Figures 7A, E ). Next, we used the more appropriate caspase 3/7 activation assay as an early marker of apoptosis which we evaluated by flow cytometry (gating strategy in Supplementary Figures 4 , 5 ). Here, we observed a notable change in the caspase 3/7 activation in both cell lines, however not statistically significant ( Figures 7B, C, F, G ). To capture both the effect of cells undergoing apoptosis and those that had already died at the time of our analysis after 48 hours of cisplatin treatment, we introduced a total dead-caspase 3/7 activation score. The score combined the percentage of caspase positive cells with percentage of dead cells calculated with the assumption that 100% of the dead cells were positive for caspase 3/7 activation during apoptosis. In IFNLR1+ conditions, we observed significantly increased total caspase 3/7 activation score when cells had been exposed to cisplatin treatment compared to mock in both cell lines ( Figures 7D, H ). In H1650, we also observed significantly higher total caspase 3/7 activation in cisplatin-treated IFNLR1+ cells compared to cisplatin-treated mock cells. Overall, these results indicate that transcriptional activation of IFNLR1 can sensitize NSCLC cell lines to cisplatin and potentiate cisplatin-induced apoptosis.