SUPPLEMENTARY MATERIAL 150 100 100 50 Supplementary Figure S1. Effect of FGFR1 and FGFR4 on lung squamous cell carcinoma lines. See also Figure 1. (a) Western blot determination of FGFR1 and FGFR4 protein expression in lung cell lines. To assess the expression of these proteins in the 18 cell lines, different blots were performed in parallel with an internal reference sample and the assembled images are shown. (b) mRNA expression determination for FGFR1 and FGFR4 in these cell lines. Clonability assays (c) and soft agar assay colony relative size representation (d) of FGFR4-overexpressing lung squamous cell carcinoma lines. Clonability (e) and soft agar assay colony relative size representation (f) of FGFR1- and FGFR4-silenced H520 cells. Colony number and size are shown for soft agar assays, and colony number is shown for clonability assays. All values were normalised to the empty vector control, and the mean of all normalised replicates is presented. Silencing of either gene was performed using two different shRNAs, referred as “a” and “b”. p-values are indicated by asterisks (* p<0.05; ** p<0.01; *** p<0.001). EV=Empty vector control, FGFR1=FGFR1-overexpressing, FGFR4=FGFR4 overexpressing, scramble=scramble shRNA control, shFGFR1=FGFR1 shRNA, shFGFR4=FGFR4 shRNA. 100 75 75 50 37 75 50 50 37 100 Supplementary Figure S2. Effect of FGFR1 and FGFR4 on lung adenocarcinoma cell lines. See also Figure 2. (a) 0.5% FBS growth curves for FGFR1- and FGFR4-overexpressing H460 cell line. (b) Clonability of FGFR1- and FGFR4-overexpressing H2009, H3122 and H460 cell lines. (c) Soft agar colony number of FGFR1- and FGFR4-overexpressing H460 cell line. (d) Soft agar assays of relative colony size for FGFR1- and FGFR4-overexpressing H2009, H3122 and H460 cell lines. (e) Western blot determination of the activation of FGFR-related signalling pathways in FGFR1– and FGFR4-overexpressing H460 cell line compared to the empty vector-containing cell line, after stimulation with FBS. (f) Soft agar assays of relative colony size for FGFR1- and FGFR4-silenced A549 cells. Silencing of either gene was performed using two different shRNAs, referred as “a” and “b”. (g) Quantification of western blots from Figure 2f. Band volume was quantified and normalised to tubulin, which was used as a loading control. Three independent replicates were quantified and normalised; median +/- SEM are shown. p-values are indicated by asterisks (* p<0.05; ** p<0.01; *** p<0.001). EV=Empty vector control, FGFR1=FGFR1-overexpressing, FGFR4=FGFR4 overexpressing, scramble=scramble shRNA control, shFGFR1=FGFR1 shRNA, shFGFR4=FGFR4 shRNA, FBS=Foetal Bovine Serum. 150 100 150 100 75 75 50 37 75 50 50 37 100 100 150 Supplementary Figure S3 Related to Figures 3 and 4. Soft agar relative colony size representation (a) and clonability assay (b) of the combination of N-cadherin overexpression and FGFR1 or FGFR4 overexpression in the lung adenocarcinoma cell lines H2009 and H3122. 0.5% FBS growth curves (c), soft agar assay (d) and western blot oncogenic signalling assessment (e) of the combination of N-cadherin silencing and FGFR1 or FGFR4 overexpression in the lung immortalised cell line NL20 with an alternative shRNA. All values in clonability and soft agar assays were normalised to the double empty vector control for each replicate of the experiment. The mean of each normalised replicate is represented. p-values are represented as asterisks (*, p<0.05; **, p<0.01; ***, p<0.001). FGFR1=FGFR1 overexpressing, FGFR4=FGFR4 overexpressing, EV1=Empty Vector 1, EV2=Empty Vector 2, CDH2=N-cadherin overexpressing, shCDH2 = N-cadherin shRNA silencing. Supplementary Figure S4. (a) Right, Correlation of expression levels of FGFR1 or FGFR4 and N-cadherin genes in GSE19188, GSE18842, GSE19804, GSE33532, GSE10072, GSE31552, TCGA, GSE2109, GSE3141, GSE14018, GSE63074 and GSE43580 lung patient databases. Data were analysed using Pearson’s r correlation. Left, Average of Pearson’s r correlation of non-tumoral (NT) or tumoral (T) GSE19188, GSE18842, GSE19804, GSE33532, GSE10072, GSE31552, TCGA, GSE2109, GSE3141, GSE14018, GSE63074 and GSE43580 lung patient databases. (b) Heatmap showing the expression levels (z-score) of FGFR1, FGFR4 and N-cadherin genes from 15 patients with the highest and 15 patients with the lowest FGFR1 or FGFR4 expression levels in four representative lung cancer patient databases. (c) Blot plots showing the N-cadherin expression levels from 15 patients with the highest and 15 patients with the lowest FGFR1 or FGFR4 expression levels in four representative lung cancer patient databases. Statistical significances (p-value) for Mann Whitney U-tests are plotted in each graph. (d) Gene set enrichment analysis (GSEA) of the FGFR1/4high-N-CDH2high transcriptional signature obtained from the TCGA lung cancer patients projected in cell lines of the CCLE grouped by FGFR1/4 and N-cadherin expression levels (FGFR1/4high-CDH2high vs FGFR1/4high-CDH2low). Supplementary Figure S5. Related to Figure 6. Data from the Cancer Cell Line Encyclopedia (CCLE) and Genomics of Drug Sensitivity in Cancer (GDSC) were analysed for an extensive panel of lung cancer cell lines treated with the FGFR1 inhibitor dovitinib (N=92, and N=63, respectively). (a) Analysis of sensitivity to the FGFR inhibitor dovitinib in lung cancer cell lines according to N-cadherin and FGFR1 or FGFR4 mRNA expression in CCLE data. Cell lines were divided according N-cadherin (CDH2) and FGFR1 or FGFR4 mRNA expression, where expression was considered to be high for values above the median expression for each gene. We found that sensitivity to dovitinib was only increased in those cell lines with high expression of both FGFR1 and N-cadherin genes. Means and standard deviations for dovitinib IC50 for each group is shown. p-values were obtained with the two-sided Mann-Whitney U test. Analysis of sensitivity to AZD4547 and PD173074 in relation to FGFR1 and CDH2 (b) or FGFR4 and CDH2 (c) expression in GDSC data. The mean expression of all probes per gene was used. Cell lines were divided according N-cadherin (CDH2) and FGFR1 or FGFR4 mRNA expression, here expression was considered to be high for values above the median expression for either FGFR, and third quartile expression for CDH2. (d) Analysis of FGFR1 amplification and MYC overexpression in the sensitivity of lung cancer cell lines in CCLE data to the FGFR inhibitor dovitinib. This figure is divided into graphs, all of which show dovitinib IC50 values, and each showing FGFR1 amplification, N-cadherin mRNA expression, FGFR1 mRNA expression and MYC mRNA expression. For the FGFR1 amplification graph, a cut-off value of > 3 was set to consider a cell line as FGFR1-amplified, with FGFR1-amplified cell lines are marked within the black boxes. In each mRNA expression graph, a coloured line indicates the median value of expression for each gene, which was set as the cut-off value to define low or high expression. Cell lines with the lowest 25% dovitinib IC50 values were considered to be sensitive. Of note, among nine cell lines with FGFR1 amplification, the most comprehensive eligibility criterion for clinical trials with FGFR inhibitors, three out of three with high N-cadherin and FGFR1 levels showed significant sensitivity to dovitinib, while five out of six cell lines with limited dovitinib sensitivity showed low N-cadherin and/or FGFR1 expression levels. In line with these results, we found FGFR inhibition sensitivity in FGFR-amplified H520 cells and in non-FGFR1-amplified H226 cells (Figure 5A). The remaining cell lines tested did not harbour FGFR1 amplification. These data were obtained from the Cancer Cell Line Encyclopedia (CCLE) database (https://portals.broadinstitute.org/ccle/home). 150 100 150 100 75 75 50 37 75 50 50 37 150 100 150 100 75 75 50 37 75 50 50 37 100 100 Figure S6. Related to Figure 7. (a) Western blot quantification of FGFR1, FGFR4 and N-cadherin expression in the PDX models under study. Band volume for these proteins was quantified and normalised to tubulin, which was used as a loading control. (b) Weight determination of AZD4547-treated mice for the treated PDX models, as indicative of AZD4547 treatment toxicity. (c) Western blot determination of the effect of AZD4547 treatment on oncogenic signalling pathways in the TP60, TP43, TP91 and TP96 PDXs models. p-values were obtained with the two-sided Mann-Whitney U test and are represented as asterisks (*, p<0.05; **, p<0.01; ***, p<0.001). SUPPLEMENTARY TABLES Supplementary Table S1. Description of the driver molecular alterations and FGFR1 amplification of our lung cell line panel Figure S1. Related to Figure 1. ADC=Adenocarcinoma, SCC=Squamous Cell Carcinoma, LCC=Large cell carcinoma, TN=“Triple Negative” (referring to the absence of alterations in KRAS, EGFR or ALK), I=Immortalised Cell line Histology Driver mutation Reference for driver mutation FGFR1 amplification (FGFR1/CEP8) A549 ADC KRAS p.G12S (Helfrich, Raben et al. 2006) 1.06 H460 LCC KRAS p.Q61H (Helfrich, Raben et al. 2006) 1.09 H2009 ADC KRAS p.G12A COSMIC 0.55 H358 ADC KRAS p.G12C (Helfrich, Raben et al. 2006) 1.38 H1650 ADC EGFR E746-E750 del (Blanco, Iwakawa et al. 2009) 0.6 H1975 ADC EGFR L858R/T790M COSMIC 0.92 HCC827 ADC EGFR E746-E750 del (Helfrich, Raben et al. 2006) 1.11 H3122 ADC EML4-ALK v1 translocation COSMIC 0.93 H2228 ADC EML4-ALK v3 translocation COSMIC 1.10 H1781 ADC TN (Helfrich, Raben et al. 2006) 0.73 H1437 ADC TN (Helfrich, Raben et al. 2006) 1.04 Calu-3 ADC TN (Helfrich, Raben et al. 2006) 2.45 (FGFR1 amplified) Calu-1 SSC KRAS p.G12C COSMIC 1.11 HTB59 SSC KRAS p-G12V COSMIC 1.13 H520 SSC TN COSMIC, (Helfrich, Raben et al. 2006) 4.04 (FGFR1 amplified) H226 SSC TN COSMIC, (Helfrich, Raben et al. 2006) 1.31 NL20 I TN COSMIC 1.55 NuLi-1 I TN COSMIC 1.07 Supplementary Table S2. Characteristics of the NSCLC cancer patient cohort Patients (n) Gender Male Female Age (years) Smoking habits Smoker Former smoker Non-smoker Squamous cell carcinoma Stage I-II Stage III Adenocarcinoma Stage I-II Stage III Other histologies 109 100 (92%) 9 (8%) 65.0 [54.0-78.0] 56 (51.3%) 48 (44.1%) 5 (4.6%) 58 (53.2%) 43 (74.1%) 15 (25.9%) 36 (33%) 30 (83.3%) 6 (16.6%) 15 (13.7%) Supplementary Table S3. Differentially expressed genes in our filtered gene expression signature for NSCLC tumours expressing high levels of FGFR1 and/or FGFR4, and high levels of N-cadherin in the TCGA RNAseq database for lung adenocarcinoma and squamous lung carcinoma. F14hhvsChl=comparison of tumours with high and low N-cadherin, considering only those tumours with high FGFR1 and/or FGFR4 expression. F14hhvsClh= comparison of tumours with high and low N-cadherin, considering only those tumours with high FGFR1 and/or FGFR4 expression. The results shown here are based in whole or in part upon data generated by the TCGA Research Network: http://cancergenome.nih.gov/. Gene F14hhvsChl.logFC F14hhvsChl.P.Value F14hhvsChl.B F14hhvsClh.logFC F14hhvsClh.P.Value F14hhvsClh.B PROX1 2,316472296 5,22E-17 28,03644776 0,520633695 0,252601838 -5,599728495 REEP2 1,645627476 2,88E-14 21,87973126 0,44583087 0,216762707 -5,494830316 RNF150 1,589519552 1,25E-13 20,44652754 0,395724466 0,269848058 -5,643918909 MLLT11 1,826922773 1,02E-12 18,40765996 0,628897241 0,14426504 -5,201569568 TUBB2B 2,669018848 1,09E-12 18,33766149 0,951149294 0,131297957 -5,131317223 AVPR1A 1,53448157 3,00E-12 17,35849631 0,130159179 0,724732396 -6,173026852 EYA2 2,254591255 1,28E-11 15,94684125 0,197813038 0,724520817 -6,172930656 TMEM74 1,646818423 1,89E-11 15,56770788 0,145775008 0,724603074 -6,172968066 SV2A 1,543591123 7,09E-11 14,28616172 0,460833169 0,250410007 -5,593846038 STRA6 2,011150741 2,45E-10 13,08375862 0,862147094 0,110310607 -4,999415032 SLITRK4 1,867411577 2,55E-10 13,04745991 0,202157373 0,686377377 -6,154139504 MMP12 2,113516801 3,22E-10 12,81960076 0,928987521 0,104092083 -4,95494437 KCNB2 1,913303023 3,55E-10 12,72643168 0,460038686 0,374319878 -5,851119389 NRXN1 2,102284226 2,06E-09 11,02386284 0,657621511 0,271322077 -5,647532318 ZFHX4 1,718168313 3,21E-09 10,59612972 0,712940197 0,150121963 -5,231008505 ZMAT4 2,181705774 4,25E-09 10,32634909 -0,375824368 0,552877799 -6,062611046 GRM4 1,934099017 5,95E-09 10,00107227 0,64744775 0,254100471 -5,603714555 KLHL14 1,51477529 7,06E-09 9,836437455 0,567475796 0,204365283 -5,453598729 TNC 1,588160968 7,51E-09 9,776634796 0,594733005 0,205411078 -5,457189803 POU3F2 2,104076853 8,01E-09 9,714497443 0,995455829 0,110682111 -5,001984088 IP6K3 1,819096141 1,03E-08 9,473358516 -0,355617537 0,51226409 -6,025230066 CHRNA1 1,581176882 1,09E-08 9,414120289 0,548552414 0,246186387 -5,582330592 GNG4 2,060855231 1,11E-08 9,400964709 -0,239418325 0,697561252 -6,159952944 FGF5 1,559057183 1,18E-08 9,343495609 0,278545025 0,550911051 -6,060917113 TMSB15A 1,569826229 1,76E-08 8,953017633 0,715637931 0,133798066 -5,145450412 INSM1 2,316325437 1,99E-08 8,838947063 0,036477767 0,958762896 -6,231738745 SOX11 2,153635026 2,00E-08 8,831062531 0,714118916 0,276961188 -5,661132382 ISL1 1,718207381 4,00E-08 8,164447527 -0,258863744 0,629016089 -6,120091686 NXPH4 1,856746244 4,03E-08 8,158527583 -0,122262824 0,832773419 -6,211458699 ST6GAL2 1,603487652 8,42E-08 7,44971104 0,822670378 0,109712589 -4,995259529 KIF5A 1,55334967 1,01E-07 7,278084709 0,538686043 0,282055201 -5,673121788 PCSK1 2,223351137 1,83E-07 6,703745948 -0,990032348 0,176904948 -5,35077785 NOL4 1,713726405 1,84E-07 6,696412703 0,326755534 0,56267026 -6,070878191 RET 1,775099002 2,86E-07 6,27575141 0,665106601 0,263670206 -5,628502228 NRCAM 1,586167227 6,91E-07 5,431822583 0,434837997 0,429450995 -5,930939232 B4GALNT4 1,577318202 1,11E-06 4,981900743 -0,347251757 0,533728293 -6,045626177 DPYSL5 1,882815979 1,36E-06 4,784498325 0,670578391 0,318606165 -5,751744625 CHRNA9 1,736372553 1,50E-06 4,693645236 -0,649339048 0,297045926 -5,706869291 IGF2 1,688476119 1,75E-06 4,544865525 0,52117931 0,392457527 -5,879163182 GALNT13 1,506639186 1,85E-06 4,49167229 0,629953826 0,248137711 -5,587680578 CDK5R2 1,560242913 2,14E-06 4,352684397 0,771139113 0,175201105 -5,343795774 ASCL1 2,607198158 2,84E-06 4,082523682 -1,080052348 0,261665578 -5,623402939 TAGLN3 1,559656288 3,04E-06 4,019066049 -0,290641281 0,614344265 -6,110172062 SCG3 1,673851438 8,18E-06 3,076533344 0,211389787 0,744648916 -6,181698725 FBN2 1,946066508 9,60E-06 2,924708996 0,820416913 0,281284694 -5,671325914 MMP3 1,768453999 1,13E-05 2,770626661 1,092096552 0,117983769 -5,050633016 CTNND2 1,89553686 1,23E-05 2,692866979 -1,100361215 0,143264361 -5,196405177 TBX18 1,602993727 1,88E-05 2,288805856 0,055896716 0,931352323 -6,229441547 PTHLH 1,823736376 1,94E-05 2,256595463 0,665085616 0,368886216 -5,84234263 SPOCK3 1,572158525 2,80E-05 1,912696381 0,163750673 0,801251785 -6,202369012 PCP4 1,748334483 6,04E-05 1,18760244 -0,358732184 0,635459188 -6,124285685 SLC6A15 1,999190858 6,31E-05 1,146328274 0,341849622 0,693555335 -6,157900199 KIF1A 2,105352509 0,000109643 0,628194341 -0,025153924 0,978782054 -6,232690193 COL2A1 1,530461482 0,000109759 0,627201243 -0,470902956 0,493617643 -6,006266275 UGT3A1 1,56859785 0,000211738 0,013045483 -0,045497438 0,950753486 -6,231186679 Supplementary Table S4. Potential upstream regulatory kinases and transcription factors of our gene signature identified by IPA using a conservative approach (p value < 0.01). The results shown here are based in whole or in part upon data generated by the TCGA Research Network: http://cancergenome.nih.gov/. Upstream Regulator Expr Log Ratio Molecule Type Activation z-score p-value of overlap ERBB3 kinase 1,45E-03 FGFR1 kinase 1,45E-03 TGFBR2 kinase 1,57E-03 NTRK1 kinase 1,68E-03 RET 1.775 kinase 3,04E-03 FGFR2 kinase 4,21E-03 MAP3K8 kinase 6,78E-03 PIK3C3 kinase 7,59E-03 MAP3K1 kinase 8,30E-03 CTNNB1 transcription regulator 2.425 6,16E-08 ASCL1 2.607 transcription regulator 1.982 3,62E-06 RBPJ transcription regulator -1.980 3,31E-05 MTF1 transcription regulator 3,42E-04 PTF1A transcription regulator 5,00E-04 NHLH2 transcription regulator 5,63E-04 REST transcription regulator 6,29E-04 EGR2 transcription regulator 1.964 7,09E-04 JUN transcription regulator 1.982 1,04E-03 TLX3 transcription regulator 1,05E-03 SCX transcription regulator 1,29E-03 EHF transcription regulator 1,50E-03 PAX3 transcription regulator 1,60E-03 GLI1 transcription regulator 1.067 1,75E-03 TAF4B transcription regulator 1,97E-03 HMGB2 transcription regulator 1,97E-03 PAX6 transcription regulator 2,01E-03 ELF3 transcription regulator 2,80E-03 MYOG transcription regulator 2,80E-03 SMAD3 transcription regulator 0,870 3,02E-03 DACH1 transcription regulator 4,19E-03 POU4F2 transcription regulator 4,19E-03 NEUROD1 transcription regulator 4,40E-03 DDX20 transcription regulator 5,06E-03 MRTFA transcription regulator 5,09E-03 PAX1 transcription regulator 6,08E-03 SMAD7 transcription regulator 6,42E-03 FOXA1 transcription regulator 7,03E-03 KLF2 transcription regulator 7,28E-03 ZNF219 transcription regulator 7,59E-03 LHX8 transcription regulator 7,59E-03 SOX2 transcription regulator 9,76E-03 image6.png image7.png image8.png image9.png image10.png image11.png image1.png image2.png image3.png image4.png image5.png