JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

1 JPET #244129

The Journal of Pharmacology and Experimental Therapeutics

The triple angiokinase inhibitor nintedanib directly inhibits tumor cell growth and induces tumor shrinkage via blocking oncogenic receptor tyrosine kinases

Frank Hilberg, Ulrike Tontsch-Grunt, Anke Baum, Anh T. Le, Robert C. Doebele, Simone

Lieb, Davide Gianni, Tilman Voss, Pilar Garin-Chesa, Christian Haslinger and Norbert Kraut

Boehringer Ingelheim RCV GmbH Co KG, Dr. Boehringer-Gasse 5-11, A-1121 Vienna, Downloaded from

Austria (F.H., U.T-G, A.B., S.L., D.G., P.G-C, C.H., N.K.)

University of Colorado, School of Medicine, Division of Medical Oncology, 12801 E. 17th jpet.aspetjournals.org

Ave., MS 8117, Aurora, CO 80045 (A.T.L., R.C.D.)

AstraZeneca - Innovative Medicines and Early Development | Discovery Sciences, 310 Unit

Darwin Building - Cambridge Science Park - Milton Rd - CB4 0WG (D.G.) at ASPET Journals on September 30, 2021

JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

2 JPET #244129

Running title: nintedanib small molecule TKI direct inhibition of tumor cell growth

Corresponding author:

Frank Hilberg

Boehringer Ingelheim RCV GmbH Co KG

Dr. Boehringer-Gasse 5-11

A-1121 Vienna, Austria

Phone: ++43-1-80105-2351 Downloaded from

Fax: ++43-1-80105-2366 e-mail: [email protected] jpet.aspetjournals.org

manuscript pages : 30 at ASPET Journals on September 30, 2021 number of tables: 3 number of figures : 4 number of references : 55

Abstract: 248 words

Introduction: 578 words

Discussion: 1200 words

ABBREVIATIONS: KD, knockdown.

Section assignment: Drug discovery and Translational Medicine

JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

3 JPET #244129

Abstract

The triple angiokinase inhibitor nintedanib is an orally available, potent and selective inhibitor of tumor angiogenesis by blocking the tyrosine kinase activities of VEGFR 1-3, PDGFR

and  and FGFR 1-3. Nintedanib has received regulatory approval in second line adenocarcinoma NSCLC in combination with docetaxel. In addition, nintedanib has been approved for the treatment of idiopathic lung fibrosis. Here we report the results from a broad kinase screen that identified additional kinases as targets for nintedanib in the low nanomolar

range. Several of these kinases are known to be mutated or overexpressed and are involved Downloaded from in tumor development (DDR1 and 2, TRKA and C, Ret) as well as in fibrotic diseases (eg.

DDRs). In tumor cell lines displaying molecular alterations in potential nintedanib targets, the inhibitor demonstrates direct anti-proliferative effects: in the NSCLC cell line NCI-H1703 jpet.aspetjournals.org carrying a PDGFR amplification; the gastric cancer cell line KatoIII and the breast cancer cell line MFM223 both driven by a FGFR2 amplification; AN3CA (endometrial carcinoma) bearing a mutated FGFR2; the AML cell lines MOLM-13 and MV-4-11-B with FLT3 mutations; at ASPET Journals on September 30, 2021 the NSCLC adenocarcinoma LC-2/ad harboring a CCDC6-RET fusion. However, potent kinase inhibition does not strictly translate into anti-proliferative activity as demonstrated in the TRKA dependent cell lines CUTO-3 and KM-12. Importantly, nintedanib treatment of

NCI-H1703 tumor xenografts triggered effective tumor shrinkage, indicating the direct effect on the tumor cells on top of the antiangiogenic effect on the tumor stroma. These findings will be instructive to guide future genome-based clinical trials with nintedanib. JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

4 JPET #244129

Introduction

The inhibition of tumor angiogenesis is an established therapeutic option in the treatment of cancer patients. Several small molecule inhibitors as well as antibodies have been approved in recent years starting with bevacizumab in 2004 (Culy, 2005) in CRC and the latest additions of ramucirumab and nintedanib in 2014 (Reinmuth et al., 2015). With the exception of renal cell carcinoma, antiangiogenic treatments are being combined with chemotherapy

(Stukalin et al., 2016) (Jayson et al., 2016) to enhance patient response and to prolong

overall survival (OS). The mechanism responsible for the enhanced efficacy of Downloaded from chemotherapeutic drugs is still not completely understood, especially since several components of the tumor stroma are involved that are not necessarily linked (Gasparini et al.,

2005a; Gasparini et al., 2005b; Boere et al., 2010; Moschetta et al., 2010; Jain, 2014). jpet.aspetjournals.org

Hypotheses include tumor vessel normalization through anti-vascular effects leading to better drug delivery and distribution (Jain, 2005; Cesca et al., 2016) and vessel co-option, the ability of tumors to hijack the existing vasculature in organs such as the lungs or liver (Kerbel, 2015). at ASPET Journals on September 30, 2021

Another possible contribution comes through the inhibition of specific ‘off-target’ kinases of antiangiogenic tyrosine kinase inhibitors (TKIs) that are often associated with the so called multi TKIs (e.g. pazopanib targeting Ret kinase). Nintedanib has been previously described as a triple angiokinase inhibitor targeting the tumor stroma and specifically the vasculature

(Hilberg et al., 2008). On the basis of the inhibition profile, nintedanib was profiled as an inhibitor of tumor angiogenesis in clinical trials that led to its regulatory approval in second line adenocarcinoma non-small cell lung cancer (NSCLC) in combination with docetaxel

(Reck et al., 2014). Nintedanib shows the strongest benefit in patients with fast progressing tumors that either do not respond to first line platinum based chemotherapy or progress within 6 months after initiation of first line chemo. This hints to a predominant antiangiogenic effect of nintedanib because such fast growing tumors heavily depend on oxygen supply and aerobic metabolism and therefore proper vascular connections (Hilberg et al J Clin Oncol 32,

2014 (suppl; abstr e22080) ASCO poster). The question of whether the extended nintedanib JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

5 JPET #244129 kinase inhibition profile can contribute to the observed clinical benefit by directly affecting tumor cell proliferation and survival remains pertinent. FGFR genetic alterations such as mutations or amplifications or fusions have been reported for the following cancers: bladder

(FGFR3) (Cancer Genome Atlas Research, 2014b), endometrial (FGFR2, (Winterhoff and

Konecny, 2016)) and lung (FGFR1, (Weiss et al., 2010; Dienstmann et al., 2014)), breast

(FGFR1 and 2), gastric (FGFR2, (Cancer Genome Atlas Research, 2014a)), lung and and glioma (FGFR3-TACC3 fusion; (Capelletti et al., 2014; Di Stefano et al., 2015)). Genetic alterations of the PDGFRA gene occur in about 5% of gastrointestinal stromal tumors (GIST) Downloaded from and amplifications are present in 5-10% of glioblastoma multiforme cases, in oligodendrogliomas, esophageal squamous cell carcinoma, artery intimal sarcomas and in 2-

3% NSCLC adenocarcinomas (Heldin, 2013). We here present data that clearly underline the jpet.aspetjournals.org potential of nintedanib to directly inhibit tumor cell proliferation and survival as single agent.

This effect of nintedanib can be seen over a wide range of tumor types and various genetic alterations ranging from mutations to amplifications and is also demonstrated in at ASPET Journals on September 30, 2021 combinations with a small molecule inhibitor targeting a tumor epigenetically. We also demonstrate that inhibition of a RTK at the kinase level does not necessarily translate into a cellular effect. Collectively, our results provide a strong rational for clinical investigations of nintedanib in specific subsets of oncogene-driven cancers.

JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

6 JPET #244129

Methods

Molecular characterization of cancer cell line panel (Ricerca 240-OncoPanel™)

Information on the Ricerca 240-OncoPanel™ can be found at: https://www.eurofinspanlabs.com/Catalog/Products/ProductDetails.aspx?prodId=YWEUPExy

%2Fhg%3D and is represented in supplementary Table S3. Relative copy number values were determined using the Affymetrix Genome-Wide Human SNP Array 6.0 platform. The analysis was performed using the CRMA v2 method (Bengtsson et al., 2009), followed by

CBS segmentation, both implemented as part of the Aroma Project (http://www.aroma- Downloaded from project.org) (Bengtsson et al., 2008) . Absolute copy numbers were calculated using the

PICNIC algorithm (http://www.sanger.ac.uk/science/tools/picnic) (Greenman et al., 2010). jpet.aspetjournals.org Mutation analysis: Mutations were called following best practices as described for instance in

(Alioto et al., 2015). RNA-seq data processing and gene expression quantification: Libraries from polyA enrichted RNA (protocols Quiagen, Illumina) were sequenced on an Illumina at ASPET Journals on September 30, 2021 HiSeq1500 using the Paired-end protocol with 50 cycles and 25 million reads in average.

Gene and transcript quantification was performed using SAMtools version 1.0 (Li et al., 2009) and Cufflinks version 2.0.2 (Trapnell et al., 2010). The molecular status of nintedanib target kinases (Table 2) in the Ricerca panel was determined as follows: Kinases with a relative copy number above 4 were considered to be amplified. The threshold for overexpression for each kinase was defined by the respective mean over all cell lines within a tumor type plus 2 times the respective standard deviation. The functional impact of mutations was derived from the MutationAssessor (http://mutationassessor.org/)(Reva et al., 2011) . Statistics calculations (Fisher test) were performed using the R open source programming language

(www.r-project.org)(R Core Team (2016) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R- project.org/).

Molecular characteristics of human tumor samples JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

7 JPET #244129

Gene expression and alteration frequencies (copy number, mutation) within the TCGA data set were analyzed using the cBio Portal online tools (www.cbioportal.org) (Cerami et al.,

2012).

Kinase assay

Kinase assays were performed at Thermo Fischer Life Technologies using standard protocols (SelectScreen® Biochemical Profiling Service). ATP concentrations used were at the apparent Km. Downloaded from Z’-LYTE® Assay Conditions: The compounds were screened in 1% DMSO (final concentration) in the well. For 10 point titrations, 3-fold serial dilutions were conducted from

the starting concentration. All Peptide/Kinase Mixtures were diluted to a 2x working jpet.aspetjournals.org concentration in the appropriate Kinase Buffer (see section Kinase Specific Assay Conditions for a complete description in supplementary information). All ATP Solutions were diluted to a

4x working concentration in Kinase Buffer (50 mM HEPES pH 7.5, 0.01% BRIJ-35, 10 mM at ASPET Journals on September 30, 2021

MgCl2, 1 mM EGTA). ATP Km apparent is previously determined using a Z´-LYTE® assay.

The Development Reagent is diluted in Development Buffer (see section Kinase-Specific

Assay Conditions - Direct and Cascade for a complete description in supplementary information). 10x Novel PKC Lipid Mix: 2 mg/ml Phosphatidyl Serine, 0.2 mg/ml DAG in 20 mM HEPES, pH 7.4, 0.3% CHAPS.

Adapta® Assay Conditions: The compounds wre screened in 1% DMSO (final concentration) in the well. For 10 point titrations, 3-fold serial dilutions are conducted from the starting concentration. All Substrate/Kinase Mixtures were diluted to a 2x working concentration in the appropriate Kinase Buffer (see section Kinase Specific Assay Conditions for a complete description in supplementary information). All ATP Solutions were diluted to a 4x working concentration in water. ATP Km apparent was previously determined using a radiometric assay except when no substrate was available in which case an Adapta® assay was conducted. The Detection Mix was prepared in TR-FRET Dilution Buffer. The Detection mix JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

8 JPET #244129 consisted of EDTA (30 mM), Eu-anti-ADP antibody (6 nM) and ADP tracer. The detection mix contained the EC60 concentration of tracer for 5-150 μM ATP.

LanthaScreen® Eu Kinase Binding Assay Conditions: The compounds were screened in 1%

DMSO (final concentration) in the well. For 10 point titrations, 3-fold serial dilutions were conducted from the starting concentration. All Kinase/Antibody Mixtures were diluted to a 2x working concentration in the appropriate Kinase Buffer (see section Kinase Specific Assay

Conditions for a complete description in supplementary information). The 4x AlexaFluor®

labeled Tracer was prepared in Kinase Buffer. Downloaded from

Cells and Western blotting

NCI-H1703 cells were seeded in 6-well plates and reverse transfected with 10nmol/l of ON- jpet.aspetjournals.org

TARGETplus Human siRNA and Lipofectamin RNAiMAX (Invitrogen #13778075) for 72h at

37°C. Transfections were performed according to the supplier´s manual using the following

siRNAs: (FGFR1 siRNA J-003131-00-0002-10; PDGFRA siRNA J-003162-00-0002-12; and at ASPET Journals on September 30, 2021

Non-targeting siRNA, D-001810-01-20), all obtained from Dharmacon. Two hours prior lysing the cells, nintedanib (BIBF 1120) and/or PD173074 were added in various concentrations to the cultures. Compounds in DMSO were diluted serially and incubated with the cells. Lysates were generated according to standard protocols (Current Protocols in Molecular Biology, edited by F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J.G. Seidman, J. A. Smith &

K. Struhl, 2004, John Wiley and Sons, Inc.). Western blotting was done using standard SDS-

PAGE methods, loading 10 µg of protein per lane, with detection by enhanced chemiluminescence. The amount of total and phoshorylated proteins were detected with the corresponding antibodies, purchased and used according to the manufacture´s instruction:

Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (Cell Signaling #9101); p44/42 MAPK

(Erk1/2) (Cell Signaling #9102 ); Akt (#9272);phosphorylated Akt (Ser473) (#4058); Cleaved

Caspase-3 (#9665); FGF Receptor 1 (D8E4) XP® Rabbit mAb (#9740); FGF Receptor 2 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

9 JPET #244129

(D4H9) (#11835) and PDGF Receptor α( #3164) were all obtained from Cell Signaling.

GAPDH was used as a loading control (ab8245, Abcam).

Cell Viability Assay

Between 3000 -10000 cells/well were plated in 96-well flat-bottom microtiter plates and incubated overnight at 37°C in a CO2 incubator with the respective medium. Test compound was added at various concentrations for 72 hours. After 6-hour incubation with Alamar Blue solution (Thermo Fischer / Invitrogen, Carlsbad, CA) at 37°C, fluorescence was measured

(Envision MultiModeReader; PerkinElmer) using an excitation wavelength of 531 nm and Downloaded from emission at 595 nm. Data Analysis: Data were fitted by an iterative calculation using a sigmoidal curve analysis program (Prism version 3.0; Graph Pad, La Jolla, CA) with variable jpet.aspetjournals.org hill slope from which EC50 values were calculated.

Incucyte Assay at ASPET Journals on September 30, 2021 Acute myeloid leukemia (AML) cells (MOLM-13, MV-4-11B, and THP-1) were plated (1 ×

104/well) in 96-well poly-d-lysine–coated black/clear bottom plates (BD BioCoat). The next day compounds were added at the indicated concentrations. Visualization and measurement of cell growth was by IncuCyte ZOOM® live cell imaging. Cell confluence was monitored for up to 140 hours during which images were taken every 3 hours.

In vivo tumor models.

Mice were housed under pathogen-free conditions (AAALAC accredited facility) and treated according to the institutional, governmental and European Union guidelines (Austrian Animal

Protection Laws, GV-SOLAS, FELASA). All animal studies were reviewed and approved by the internal ethics committee and the local governmental committee. Six to eight-week-old female BomTAC:NMRI-Foxn1nu mice (21-31 g) were inoculated with 1x 106 (in 100 µl) NCI-

H1703 or 2.5 x 106 MV-4-11-B cells s.c. into the right flank. Tumors were measured three times weekly using calipers. Volumes were calculated according to the formula “tumor JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

10 JPET #244129 volume = length * diameter2 * π/6.” Tumor growth inhibition (TGI) was calculated to the formula: “TGI =100 x {1-[(treatedfinal day– treatedday1) / (controlfinal day– controlday1)]}”

One-sided decreasing Mann–Whitney tests were used to compare tumor volumes (efficacy).

The P values for the tumor volume assessment were adjusted for multiple comparisons according to Bonferroni–Holm. For all analyses, P values under 0.05 represented a statistically significant effect. Kaplan-Meier survival curves were calculated calculation using a curve analysis program (Prism version 3.0; Graph Pad, La Jolla, CA).

Downloaded from

jpet.aspetjournals.org

at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

11 JPET #244129

Results

Nintedanib shows potent inhibition of FGFR tyrosine kinases 1-3 and PDGFR tyrosine kinases 

A head-to-head comparison of nintedanib to in-class competitor molecules revealed that nintedanib is more potent and has a more balanced inhibition profile across the triple angiokinase panel. Nintedanib was compared to sunitinib, vandetanib, pazopanib and cediranib on the basis of the IC50 values determined in parallel assays (Table 1). For Downloaded from nintedanib the previously reported potency on VEGFR1-3, PDGFR and  and on FGFR1-3 could be replicated with IC50 values below 100nmol/l with slightly lower IC50 values for

VEGFR2 and 3 and PDGFR and  (Hilberg et al., 2008). Sunitinib, vandetanib, pazopanib jpet.aspetjournals.org and cediranib inhibit VEGFR-2 and -3 at concentrations below 100nmol/l, but their inhibition of PDGFR  and  is less potent and there is a clear potency drop with respect to FGFR inhibition. The notable exception is cediranib which exhibits inhibition across the FGFRs at ASPET Journals on September 30, 2021 which is comparable to nintedanib. In this comparison, sorafenib is the least potent drug with

IC50 values equal to or higher than 100nmol/l across the tested kinase panel.

Nintedanib potently inhibits oncogenic RTKs beyond FGFRs and PDGFRs

The previously reported kinase target spectrum presented nintedanib as a triple angiokinase small molecule TKI (Hilberg et al., 2008). To explore kinase inhibition beyond that described previously, nintedanib was tested on a much broader, 245 kinase panel. In addition to the 13 kinases that had been previously reported as nintedanib targets, we identified 21 additional kinases that are also inhibited by nintedanib with IC50 values below 100nmol/l (Table 2).

Among these newly identified targets are kinases well known to be deregulated through genetic alterations and thus may act as oncogenic drivers in human cancers. These include kinases altered in subsets of NSCLC such as Ret (2nmol/l; (Berge and Doebele, 2014) and

TRKA (35 nmol/l; (Marchetti et al., 2008). The kinases Abl1 and Kit (13 and 9 nmol/l, respectively) are also known to be driving mutations in several tumour types including JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

12 JPET #244129 leukemias (Greuber et al., 2013) (Stankov et al., 2014). Nintedanib also inhibits DDR1 and 2

(23 and 18 nmol/l, respectively) which have been described as being involved in inflammatory and fibrotic processes (Guerrot et al., 2011; Olaso et al., 2011).

The fact that TKIs interfere with kinase activity in in vitro kinase assays does not necessarily mean that this activity will translate into cellular activity. We addressed this issue in two cell lines with known rearrangements of NTRK1 leading to constitutive TRKA receptor activation: the CUTO-3.29 NSCLC adenocarcinoma cell line bearing the MPRIP-NTRK1 rearrangement

and (Doebele et al., 2015)) and the microsatellite instability (MSI) high colorectal cancer Downloaded from

(CRC) cell line KM-12 which has the TPM3-NTRK1 translocation (Camps et al., 2004). As shown in Figure 1, the CUTO-3.29 (Figure 1C) and KM12 (Figure 1D) cells have high levels of pTRKA resulting in elevated pERK1/2 and pAKT levels. As demonstrated, these pTRKA jpet.aspetjournals.org and pERK1/2 levels can be reduced by the addition of the TRK-specific inhibitor, entrectinib, at concentrations as low as 1 nmol/l. In contrast, much higher concetrations of nintedanib at ASPET Journals on September 30, 2021 (≥1000 nmol/l) were required to reduce pTRKA and pERK1/2 levels. The EC50 values of nintedanib and entrectinib were calculated to be 955 nmol/l and 117 nmol/l for the CUTO-

3.29 cells and 557 nmol/l and 2.3 nmol/l for the KM12 cell line, respectively (Figure 1A and

B). Thus, although nintedanib is a relatively potent inhibitor of TRKA in the kinase panel, this potency does not translate into inhibition of receptor phosphorylation and downstream signaling in cancer cell lines bearing TRKA rearrangements.

Nintedanib inhibits tumor cell proliferation directly

In order to address the cellular inhibitory capacity of nintedanib in more detail the drug was tested on a large collection of cell lines (Ricerca 240-OncoPanel™, supplementary Table S3).

Nintedanib displayed efficacy (cutoff: 500 nmol/l, supplementary Table S1a) against 12 of the

242 cancer cell lines across several indications (Figure 2A) including gastric carcinoma, chronic myelogenous leukemia (CML), acute myeloid leukemia (AML), NSCLC, renal cancer, rhabdomyosarcoma and thyroid carcinoma (Supplementary Table S1b). These cell lines JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

13 JPET #244129 carry driver mutations that may, potentially, be targeted by nintedanib. For example, the AML cell line MV-4-11 expresses the mutated FLT3 ITD allele exclusively which acts as the oncogenic driver (Quentmeier et al., 2003) and which is inhibited by nintedanib with a GI50 of

53 nmol/l (supplementary Table S1a). Of the 12 sensitive cell lines, 10 exhibited genetic alterations within a total number of 16 nintedanib target kinases, either as mutation, amplification or overexpression (Figure 2B, Table S2). The two sensitive cell lines that do not carry mutations in any of the nintedanib target kinases are the highly mutated A-427 (NSCLC) line which carryies 418 mutations and the RCC cell line G-401 with 311 mutations (Table Downloaded from S1a). In addition, the A-427 cell line shows a highly complex amplification pattern. The 16 kinase genes as shown in Fig. 2B are altered to a similar extent in both sensitive and insensitive cell lines. jpet.aspetjournals.org

The TCGA data sets for those tumor types represented by the sensitive cell lines was queried to determine the total frequency of alterations in the 16 nintedanib targetable kinases.

Around 42 % of gastric cancer cases harbored molecular alterations in any of the described at ASPET Journals on September 30, 2021

16 genes. The alterations detected in Acute Myeloid Leukemia and Chronic Lymphatic

Leukemia were mostly mutations compared to amplifications or rearrangements

(supplementary Figure S1). The kinases overexpressed in the nintedanib sensitive cell lines were, to a certain extent, also overexpressed in the respective TCGA data sets

(supplementary Figure S2). In summary, 8 of the 12 nintedanib sensitive cell lines from the

242 cancer panel contained genetic alterations in kinases targeted by nintedanib.

Effects on growth of tumor cell lines in vitro where nintedanib targets may be driver mutations

We determined the responses of the the NSCLC cell line NCI-H1703 (PDGFR and FGFR1 ampl.), the gastric cancer line KatoIII (FGFR2 ampl.), the endometrial cancer cell line AN3CA

(FGFR2 mut.) and the breast cancer cell line MFM-223 (FGFR2 ampl.) to treatment with nintedanib. As shown in Table 1B, growth of all 4 cell lines was inhibited by nintedanib JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

14 JPET #244129

treatment with EC50 values in the lower nanomolar range: NCI-H1703 10 nmol/l, KatoIII 176 nmol/l, AN3CA 152 nmol/l and MFM-223 108 nmol/l. As expected, sorafenib which is a very weak inhibitor of PDGFR and FGFR kinase activity (see Table 1A) was also a weak inhibitor of the NCI-H1703 and Kato III cell lines (EC50 values of 258 nmol/l and 383 nmol/l respectively; Table 1B). Sunitinib, which had IC50 values of 71 nmol/l and 184 nmol/l against

PDGFR and FGFR2 kinases respectively (Table 1A) was also markedly less potent than nintedanib with EC50 values of 39 nmol/l for the NCI-H1703 cells and 624 nmol/l against the

KatoIII line (Table 1B). The differences between nintedanib, sunitinib and sorafenib were Downloaded from confirmed by signaling pathway analysis in NCI-H1703 cells following PDGF BB stimulation.

As shown in Figure 3D, nintedanib and sunitinib were able to reduce pMAPK, pAKT and

pPDGFR levels in a concentration-dependent manner down to 10-50 nmol/l whereas jpet.aspetjournals.org sorafenib was only able to fully interfere with MAPK, AKT and PDGFR activation at the highest tested concentration of 1 µmol/l. at ASPET Journals on September 30, 2021 Figure 3A also shows that activation of MAPK after stimulation with PDGF BB is inhibited by nintedanib at concentrations of 10 nmol/l and above. The same effect can be seen for phosphorylated AKT. In addition the conversion of PARP to cleaved PARP as a marker of apoptosis is also seen at nintedanib concentrations 30 nmol/l and above. In comparison we also examined the effects of imatinib a TKI that weakly inhibits PDGFR (Capdeville et al.,

2002) in this setting. As shown in Figure 3A, imatinib interferes with MAPK and AKT activation only at the highest concentration tested (1,000 nmol/l) resulting in a faint cPARP signal. This difference between nintedanib and imatinib results from the lower (10-fold) potency of imatinib in inhibiting PDGFRB kinase activity which is corroborated by the EC50 values of imatinib in comparison to those of nintedanib shown in (Table 1B).

As shown in supplementary Figure S3, both the NCI-H1703 and KatoIII cell lines have two genetic amplifications. PDGFRA and FGFR1 are focally amplified in the NCI-H1703 cells whereas FGFR2 and PDGFRA are amplified in the KatoIII cells. In order to determine which genetic alteration is driving proliferation and survival in these two cell lines, knockdown JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

15 JPET #244129 experiments were performed. Figure 3B shows that knocking down FGFR1 in the NCI-H1703 cells had no effect on pAKT and pMAPK whereas the PDGFRA knock down (KD) reduced the pMAPK as well as the pAKT signal. The addition of increasing concentrations of nintedanib to the PDGFRA KD resulted in a complete loss of the pMAPK and pAKT signals.

The addition of nintedanib to the FGFR1 KD also had an impact on the pAKT and MAPK signals but to a lesser extent demonstrating that NCI-H1703 cells are primarily driven by the

PDGFRA amplification.

Among the cancer indications with nintedanib-targetable driver mutations, gastric cancer Downloaded from shows a FGFR2 amplification prevalence of 4-10% (Cancer Genome Atlas Research, 2014a).

As a representative from this group we analysed the effects of nintedanib on proliferation and survival pathways in the KatoIII cells. This cell line has been reported to have a copy number jpet.aspetjournals.org gain of 17 for FGFR2. Nintedanib inhibited proliferation in this cell line with an IC50 of 176 nmol/l and interfered with MAPK and AKT activation in bFGF stimulated KatoIII cells down to

100 nmol/l and 30 nmol/l, respectively (Figure 3C). In the same experiment, the pan-FGFR at ASPET Journals on September 30, 2021 inhibitor PD173074 inhibited KatoIII cell proliferation with an IC50 of 30 nmol/l and interfered with MAPK and AKT at a concentration as low as 10 nmol/l.

The NSCLC adenocarcinoma cell line LC-2/ad which carries a CCDC6-RET fusion as a driver mutation (Suzuki et al., 2013) was also tested for its response to nintedanib, based on our finding that RET is also a target for the drug (EC50 value = 149 nmol/l). In the same experiment vandetanib, a marketed drug for the treatment of thyroid carcinoma based on its

Ret inhibition, gave an EC50 value of 247 nmol/l (data not shown).

Based on the previously reported inhibition of Flt3 by nintedanib (Kulimova et al., 2006;

Hilberg et al., 2008) we were interested in exploring the combination with the small molecule

BET (bromodomain and extraterminal) family inhibitor BI 894999 (Effects of the novel BET inhibitor BI 894999 on upregulation of HEXIM1 in cancer cells and on anti-tumor activity in xenograft tumor models. J Clin Oncol 34, 2016, suppl; abstr 11574 (ASCO)). As JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

16 JPET #244129 demonstrated in supplementary Figure 4, nintedanib specifically contributes to the inhibition of proliferation in Flt3 mutated AML cell lines (MOLM-13 and MV-4-11-B) but did not inhibit the proliferation of THP-1 AML cells carrying wild type Flt3. The combination of nintedanib with BI 894999 completely inhibited the proliferation of MOLM-13 and MV-4-11-B AML cells, whereas there was no beneficial effect of the combination in the Flt3 wild type cell line

(Fiskus et al., 2014).

In vivo anti-tumor efficacy of nintedanib in tumor xenografts bearing nintedanib-

targeted driver alterations Downloaded from

As nintedanib potently inhibited in vitro growth of the NCI-H1703 NSCLC cell line which bears a PDGFRα amplification we decided to analyze the in vivo anti-tumor efficacy in tumor jpet.aspetjournals.org bearing mice. As demonstrated in Figure 4A and B, mice bearing subcutaneous NCI-H1703 tumors were treated with either nintedanib (blocking VEGFR-2 and PDGFRs), vatalanib, a predominantly VEGFR-2 inhibiting TKI, and two different doses of the PDGFR inhibitor at ASPET Journals on September 30, 2021 imatinib (75 mg/kg qd and bid). The selection of the doses of the drugs for the in vivo experiment was based on previously published reports demonstrating exposure of >1µmol/l peak plasma concentrations with the selected doses of 100 mg/kg for nintedanib (Hilberg et al., 2008) and vatalanib (Wood et al., 2000). In accordance with its in vitro activity on NCI-

H1703 cells nintedanib was the most potent of the 3 compounds in terms of its in vivo anti- tumor efficacy resulting in tumor shrinkage with a TGI value of 107% at 100 mg/kg. In comparison, the PDGFR inhibitor imatinib at 75mg/kg either once or twice daily achieved TGI values of 45% and 58%, respectively. Once daily treatment with vatalanib at a dose of

100mg/kg resulted in a TGI of 73%, which is in line with effects observed for anti-angiogenic drugs. All treatments were well tolerated as demonstrated by the weight gain of all treatment groups over the treatment period (data not shown).

The anti-tumor efficacy of nintedanib alone and in combination with the BET family inhibitor

BI 894999 was assessed in the subcutaneous MV-4-11-B AML model as shown in Figure 4C, JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

17 JPET #244129

D and E. The TGI values for the single agent BET inhibitor BI 894999 were 78%, 92% for nintedanib and 99% for the combination. The combination was well tolerated and dosing could be maintained for almost 100 days as also shown in the Kaplan-Meyer plot in Figure

4E. Seven out of eight animals in the combination treatment group survived until the end of the experiment with only one treatment unrelated death. In the BI 894999 treatment group all animals had to be euthanized according to termination criteria at around day 50 and in the nintedanib group five animals were euthanized at day 100.

Downloaded from

jpet.aspetjournals.org at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

18 JPET #244129

Discussion

The triple angiokinase inhibitor nintedanib has received regulatory approval for the treatment of idiopathic lung fibrosis as well as for second line NSCLC in combination with docetaxel

(Bronte et al., 2016). In this report we present, for the first time, pharmacological evidence that nintedanib directly and potently inhibits growth of tumors that are driven by genetic alterations in nintedanib target kinases. A screen of 240 tumor cell lines revealed 12 nintedanib- sensitive cell lines, nearly half of which were derived from leukemias. The 16 nintedanib-sensitive kinases were either amplified, mutated or overexpressed in these cell Downloaded from lines. None of the mutations represent known functional hotspots in the respective kinase domains and their functional impact is, as yet, unknown. However, the activating nature of mutations outside the kinase domain has for example been demonstrated for PI3 kinase jpet.aspetjournals.org

(Kang et al., 2005). The 2 cell lines that did not exhibit mutations in the nintedanib target genes showed multiple mutations and amplification patterns which were too complex to derive clear sensitivity determinants for nintedanib. Alterations to the 16 nintedanib-sensitive at ASPET Journals on September 30, 2021 kinases are not exclusive to the sensitive cell lines, they are also present in the insensitive cell lines which we assume depend on other alterations and combinations to drive malignant behavior.

The TCGA database was interrogated for a combined aberration incidence for these kinases.

The highest percentage (42.2%) was observed for gastric cancer, followed by AML (10 %).

FLT4 and PDGFRB are frequently overexpressed in gastric cancer and AML indicating that nintedanib might be useful in the treatment of these particular patient populations.

This is potentially important in a scientific / clinical environment where tumor treatment is becoming increasingly fragmented based on the genetic information acquired through analysis of pre-treatment tumor biopsies. This approach is now well established in the treatment of NSCLC patients bearing EML-ALK or EGFR mutations (Rocco et al., 2016), who receive appropriate TKIs targeting these alterations in the first line setting. Mutational screening of cancer is increasingly becoming incorporated into standard clinical practice requiring an increased need for targeted and well tolerated drugs. As demonstrated in this JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

19 JPET #244129 report, beyond potently inhibiting tumor angiogenesis, nintedanib also effectively blocks proliferation of tumor cells driven by alterations in PDGFRA, FGFR2, RET and FLT3. We show that this anti-proliferative activity also translates into robust single-agent in vivo anti- tumor efficacy upon treatment with nintedanib, including tumor shrinkage.

These results show that nintedanib has a dual mechanism: on the one hand, it suppresses tumor angiogenesis and on the other it interferes directly with tumor cell proliferation, resulting in the induction of apoptosis. This opens up new treatment possibilities for nintedanib in tumor subtypes depending on PDGFRA, FGFR2, RET or FLT3 genetic Downloaded from alterations either alone or in combination with chemotherapy. AML associated with an internal tandem duplication of the FLT3 gene (FLT3 ITD mutation) has been reported to occur in about 20% of AML patients (Dohner et al., 2015). It is associated with unfavorable jpet.aspetjournals.org outcome and patients are likely to relapse. A new class of compounds, BET inhibitors, are being intensively investigated preclinically and in phase I trials in AML. Combining the BET inhibitor BI 894999 with nintedanib may lead to improved survival benefit with good at ASPET Journals on September 30, 2021 tolerability in the FLT3 mutant subgroup of AML. These findings extend a previous report on combining a BET inhibitor with a FLT3 tyrosine kinase inhibitor in an in vitro setting (Fiskus et al., 2014).

In clinical practice nintedanib has demonstrated good tolerability and easy patient management. Here we show that nintedanib is a potent and well-balanced triple angiokinase inhibitor in comparison to other in-class competitors. Across the board of VEGFR, PDGFR and FGFR isoforms constituting the triple angiokinase inhibition profile, nintedanib exhibits low nanomolar IC50 values whereas competitors such as sunitinib, vandetanib, pazopanib and sorafenib completely lack FGFR activity and are less potent inhibitors of VEGFR1-3 and

PDGFR  and . Despite extending the kinase inhibition profile to 34 kinases as targets for nintedanib, the drug is well tolerated by patients on a daily dosing schedule as compared with in-class drugs that require intermittent dosing regimens to minimize adverse effects (Lee and Motzer, 2015). Furthermore, nintedanib has not been associated with serious side effects such as hand-foot syndrome (Li et al., 2015). JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

20 JPET #244129

More FGFR selective TKIs (e.g. AZD4547, BGJ398) in cancer treatment have demonstrated promising clinical efficacies in phase 2 clinical trials but are also associated with notable adverse effects such as hand-foot syndrome and serous retinopathy (Katoh, 2016). Hibi et al. have addressed the effects of nintedanib in FGFR1 amplified human lung squamous cell carcinoma cell lines H520 and LK-2 and could demonstrate that nintedanib inhibited the proliferation of FGFR1 CNG-positive LSCC cell lines in association with attenuation of the

FGFR1–ERK signaling pathway in vitro and in vivo (Hibi et al., 2016). These data, together with the results presented in this manuscript demonstrate the potential of nintedanib as a Downloaded from potent inhibitor of cancer cells bearing FGFR alterations (amplifications, mutations) that warrants confirmation in clinical trials investigating nintedanib in cancer patients whose tumors harbor FGFR alterations. jpet.aspetjournals.org

Nintedanib’s dual activity as a TKI targeting both genetic alterations in tumor cells (FGFRs and PDGFRs) as well as the tumor stroma by interfering with tumor angiogenesis may represent an advantage over drugs that target specific RTKs more selectively. The at ASPET Journals on September 30, 2021 importance of normalizing the tumor stromal environment becomes ever more important in the context of the up-coming combinations with immune-checkpoint inhibitors (Hendry et al.,

2016). Emerging strong data underscore the importance to improve the immune response by interfering with the VEGF driven negative effects on dendritic cell maturation, T-cell activation, regulatory T cell (Treg) proliferation and promotion of myeloid-derived suppressor cells

(MDSCs), as well as interfering with the activated tumor vessels and normalizing the tumor vessel bed (Hendry et al., 2016).

The fact that 34 kinases are inhibited at concentrations below 100 nmol/l identifies nintedanib as a multi TK inhibitor. But this fact is somewhat misleading because mere kinase inhibition does not always translate into inhibitory activity in a cellular context. In this report we clearly showed that the kinase inhibition of NTRK1 with an IC50 value of 30 nmol/l did not translate into the cellular inhibition of cell lines harboring a TRKA receptor activation. The same discrepancy between kinase and cellular activity has already been observed for LYN and JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

21 JPET #244129

LCK kinases with a much weaker cellular activity observed as would have been predicted from the kinase inhibition data (Hilberg et al., 2008).

Nintedanib has demonstrated to be a potent antiangiogenic and antifibrotic drug that has received regulatory approval as an anticancer agent and as an antifibrotic drug for the treatment of second line NSCLC in combination with docetaxel and idiopathic lung fibrosis, respectively. The additional features described in this report, namely the anti-proliferative activity in tumors driven by PDGFR, FGFR, FLT3 and RET alterations, can be seen as a future added value for nintedanib’s clinical applications in cancer treatment. Similar Downloaded from approaches are already implemented in clinical practice for lung tumors driven by EML-ALK rearrangements (Blackhall and Cappuzzo, 2016) or RET mutated thyroid carcinomas for crizotinib and vandetanib respectively (Chu et al., 2012). Nintedanib is already being tested jpet.aspetjournals.org clinically in some of these sub-indications with data becoming available in the coming years

(e.g. NCT02278978, NCT02299141).

at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

22 JPET #244129

Acknowledgements

We are grateful to Monika Kriz, Matthew Kennedy, Martina Scherer, Regina Ruzicka, Nina

Rodi, Susanne Schmittner and Susy Straubinger for excellent technical assistance and data evaluation and to Dr. Elizabeth Anderson for critical reading the manuscript. Downloaded from jpet.aspetjournals.org at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

23 JPET #244129

Authorship Contribution:

Participated in the research design: Hilberg, Tontsch-Grunt, Baum, Le, Doebele, Lieb, Gianni,

Voss, Garin-Chesa, Haslinger, Kraut

Conducted experiments: Tontsch-Grunt, Baum, Le, Lieb, Gianni, Voss

Performed data analysis: Hilberg, Tontsch-Grunt, Baum, Le, Doebele, Lieb, Gianni, Voss,

Garin-Chesa, Haslinger

Wrote or contributed to the writing of the manuscript: Hilberg, Tontsch-Grunt, Baum, Le, Downloaded from

Doebele, Lieb, Gianni, Voss, Garin-Chesa, Haslinger, Kraut jpet.aspetjournals.org at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

24 JPET #244129

References

Alioto TS, Buchhalter I, Derdak S, Hutter B, Eldridge MD, Hovig E, Heisler LE, Beck TA, Simpson JT, Tonon L, Sertier AS, Patch AM, Jager N, Ginsbach P, Drews R, Paramasivam N, Kabbe R, Chotewutmontri S, Diessl N, Previti C, Schmidt S, Brors B, Feuerbach L, Heinold M, Grobner S, Korshunov A, Tarpey PS, Butler AP, Hinton J, Jones D, Menzies A, Raine K, Shepherd R, Stebbings L, Teague JW, Ribeca P, Giner FC, Beltran S, Raineri E, Dabad M, Heath SC, Gut M, Denroche RE, Harding NJ, Yamaguchi TN, Fujimoto A, Nakagawa H, Quesada V, Valdes-Mas R, Nakken S, Vodak D, Bower L, Lynch AG, Anderson CL, Waddell N, Pearson JV, Grimmond SM, Peto M, Spellman P, He M, Kandoth C, Lee S, Zhang J, Letourneau L, Ma S, Seth S, Torrents D, Xi L, Wheeler DA, Lopez-Otin C, Campo E, Campbell PJ, Boutros PC, Puente XS, Gerhard DS, Pfister SM, McPherson JD, Hudson TJ, Schlesner M, Lichter P, Eils R, Jones DT and Gut IG (2015) A comprehensive assessment of somatic mutation detection in cancer using whole- genome sequencing. Nat Commun 6:10001. Bengtsson H, Irizarry R, Carvalho B and Speed TP (2008) Estimation and assessment of raw copy Downloaded from numbers at the single locus level. Bioinformatics 24:759-767. Bengtsson H, Wirapati P and Speed TP (2009) A single-array preprocessing method for estimating full-resolution raw copy numbers from all Affymetrix genotyping arrays including GenomeWideSNP 5 & 6. Bioinformatics 25:2149-2156. Berge EM and Doebele RC (2014) Targeted therapies in non-small cell lung cancer: emerging jpet.aspetjournals.org oncogene targets following the success of epidermal growth factor receptor. Semin Oncol 41:110-125. Blackhall F and Cappuzzo F (2016) Crizotinib: from discovery to accelerated development to front-line treatment. Ann Oncol 27 Suppl 3:iii35-iii41. Boere IA, Hamberg P and Sleijfer S (2010) It takes two to tango: combinations of conventional at ASPET Journals on September 30, 2021 cytotoxics with compounds targeting the vascular endothelial growth factor-vascular endothelial growth factor receptor pathway in patients with solid malignancies. Cancer Sci 101:7-15. Bronte G, Passiglia F, Galvano A, Barraco N, Listi A, Castiglia M, Rizzo S, Fiorentino E, Bazan V and Russo A (2016) Nintedanib in NSCLC: evidence to date and place in therapy. Ther Adv Med Oncol 8:188-197. Camps J, Morales C, Prat E, Ribas M, Capella G, Egozcue J, Peinado MA and Miro R (2004) Genetic evolution in colon cancer KM12 cells and metastatic derivates. Int J Cancer 110:869-874. Cancer Genome Atlas Research N (2014a) Comprehensive molecular characterization of gastric adenocarcinoma. Nature 513:202-209. Cancer Genome Atlas Research N (2014b) Comprehensive molecular characterization of urothelial bladder carcinoma. Nature 507:315-322. Capdeville R, Buchdunger E, Zimmermann J and Matter A (2002) Glivec (STI571, imatinib), a rationally developed, targeted anticancer drug. Nat Rev Drug Discov 1:493-502. Capelletti M, Dodge ME, Ercan D, Hammerman PS, Park SI, Kim J, Sasaki H, Jablons DM, Lipson D, Young L, Stephens PJ, Miller VA, Lindeman NI, Munir KJ, Richards WG and Janne PA (2014) Identification of recurrent FGFR3-TACC3 fusion oncogenes from lung adenocarcinoma. Clin Cancer Res 20:6551-6558. Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, Jacobsen A, Byrne CJ, Heuer ML, Larsson E, Antipin Y, Reva B, Goldberg AP, Sander C and Schultz N (2012) The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2:401-404. Cesca M, Morosi L, Berndt A, Fuso Nerini I, Frapolli R, Richter P, Decio A, Dirsch O, Micotti E, Giordano S, D'Incalci M, Davoli E, Zucchetti M and Giavazzi R (2016) Bevacizumab-Induced Inhibition of Angiogenesis Promotes a More Homogeneous Intratumoral Distribution of Paclitaxel, Improving the Antitumor Response. Mol Cancer Ther 15:125-135. JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

25 JPET #244129

Chu CT, Sada YH and Kim ES (2012) Vandetanib for the treatment of lung cancer. Expert Opin Investig Drugs 21:1211-1221. Culy C (2005) Bevacizumab: antiangiogenic cancer therapy. Drugs Today (Barc) 41:23-36. Di Stefano AL, Fucci A, Frattini V, Labussiere M, Mokhtari K, Zoppoli P, Marie Y, Bruno A, Boisselier B, Giry M, Savatovsky J, Touat M, Belaid H, Kamoun A, Idbaih A, Houillier C, Luo FR, Soria JC, Tabernero J, Eoli M, Paterra R, Yip S, Petrecca K, Chan JA, Finocchiaro G, Lasorella A, Sanson M and Iavarone A (2015) Detection, Characterization, and Inhibition of FGFR-TACC Fusions in IDH Wild-type Glioma. Clin Cancer Res 21:3307-3317. Dienstmann R, Rodon J, Prat A, Perez-Garcia J, Adamo B, Felip E, Cortes J, Iafrate AJ, Nuciforo P and Tabernero J (2014) Genomic aberrations in the FGFR pathway: opportunities for targeted therapies in solid tumors. Ann Oncol 25:552-563. Doebele RC, Davis LE, Vaishnavi A, Le AT, Estrada-Bernal A, Keysar S, Jimeno A, Varella-Garcia M, Aisner DL, Li Y, Stephens PJ, Morosini D, Tuch BB, Fernandes M, Nanda N and Low JA (2015) An Oncogenic NTRK Fusion in a Patient with Soft-Tissue Sarcoma with Response to the

Tropomyosin-Related Kinase Inhibitor LOXO-101. Cancer Discov 5:1049-1057. Downloaded from Dohner H, Weisdorf DJ and Bloomfield CD (2015) Acute Myeloid Leukemia. N Engl J Med 373:1136- 1152. Fiskus W, Sharma S, Qi J, Shah B, Devaraj SG, Leveque C, Portier BP, Iyer S, Bradner JE and Bhalla KN (2014) BET protein antagonist JQ1 is synergistically lethal with FLT3 tyrosine kinase inhibitor

(TKI) and overcomes resistance to FLT3-TKI in AML cells expressing FLT-ITD. Mol Cancer Ther jpet.aspetjournals.org 13:2315-2327. Gasparini G, Longo R, Fanelli M and Teicher BA (2005a) Combination of antiangiogenic therapy with other anticancer therapies: results, challenges, and open questions. J Clin Oncol 23:1295- 1311. Gasparini G, Longo R, Toi M and Ferrara N (2005b) Angiogenic inhibitors: a new therapeutic strategy

in oncology. Nat Clin Pract Oncol 2:562-577. at ASPET Journals on September 30, 2021 Greenman CD, Bignell G, Butler A, Edkins S, Hinton J, Beare D, Swamy S, Santarius T, Chen L, Widaa S, Futreal PA and Stratton MR (2010) PICNIC: an algorithm to predict absolute allelic copy number variation with microarray cancer data. Biostatistics 11:164-175. Greuber EK, Smith-Pearson P, Wang J and Pendergast AM (2013) Role of ABL family kinases in cancer: from leukaemia to solid tumours. Nat Rev Cancer 13:559-571. Guerrot D, Kerroch M, Placier S, Vandermeersch S, Trivin C, Mael-Ainin M, Chatziantoniou C and Dussaule JC (2011) Discoidin domain receptor 1 is a major mediator of inflammation and fibrosis in obstructive nephropathy. Am J Pathol 179:83-91. Heldin CH (2013) Targeting the PDGF signaling pathway in tumor treatment. Cell Commun Signal 11:97. Hendry SA, Farnsworth RH, Solomon B, Achen MG, Stacker SA and Fox SB (2016) The Role of the Tumor Vasculature in the Host Immune Response: Implications for Therapeutic Strategies Targeting the Tumor Microenvironment. Front Immunol 7:621. Hibi M, Kaneda H, Tanizaki J, Sakai K, Togashi Y, Terashima M, De Velasco MA, Fujita Y, Banno E, Nakamura Y, Takeda M, Ito A, Mitsudomi T, Nakagawa K, Okamoto I and Nishio K (2016) FGFR gene alterations in lung squamous cell carcinoma are potential targets for the multikinase inhibitor nintedanib. Cancer Sci 107:1667-1676. Hilberg F, Roth GJ, Krssak M, Kautschitsch S, Sommergruber W, Tontsch-Grunt U, Garin-Chesa P, Bader G, Zoephel A, Quant J, Heckel A and Rettig WJ (2008) BIBF 1120: triple angiokinase inhibitor with sustained receptor blockade and good antitumor efficacy. Cancer Res 68:4774- 4782. Jain RK (2005) Antiangiogenic therapy for cancer: current and emerging concepts. Oncology (Williston Park) 19:7-16. Jain RK (2014) Antiangiogenesis strategies revisited: from starving tumors to alleviating hypoxia. Cancer Cell 26:605-622. JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

26 JPET #244129

Jayson GC, Kerbel R, Ellis LM and Harris AL (2016) Antiangiogenic therapy in oncology: current status and future directions. Lancet 388:518-529. Kang S, Bader AG and Vogt PK (2005) Phosphatidylinositol 3-kinase mutations identified in human cancer are oncogenic. Proc Natl Acad Sci U S A 102:802-807. Katoh M (2016) FGFR inhibitors: Effects on cancer cells, tumor microenvironment and whole-body homeostasis (Review). Int J Mol Med 38:3-15. Kerbel RS (2015) A Decade of Experience in Developing Preclinical Models of Advanced- or Early- Stage Spontaneous Metastasis to Study Antiangiogenic Drugs, Metronomic Chemotherapy, and the Tumor Microenvironment. Cancer J 21:274-283. Kulimova E, Oelmann E, Bisping G, Kienast J, Mesters RM, Schwable J, Hilberg F, Roth GJ, Munzert G, Stefanic M, Steffen B, Brandts C, Muller-Tidow C, Kolkmeyer A, Buchner T, Serve H and Berdel WE (2006) Growth inhibition and induction of apoptosis in acute myeloid leukemia cells by new indolinone derivatives targeting fibroblast growth factor, platelet-derived growth factor, and vascular endothelial growth factor receptors. Mol Cancer Ther 5:3105-

3112. Downloaded from Lee CH and Motzer RJ (2015) Sunitinib as a paradigm for tyrosine kinase inhibitor development for renal cell carcinoma. Urol Oncol 33:275-279. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R and Genome Project Data Processing S (2009) The Sequence Alignment/Map format and

SAMtools. Bioinformatics 25:2078-2079. jpet.aspetjournals.org Li Y, Gao ZH and Qu XJ (2015) The adverse effects of sorafenib in patients with advanced cancers. Basic Clin Pharmacol Toxicol 116:216-221. Marchetti A, Felicioni L, Pelosi G, Del Grammastro M, Fumagalli C, Sciarrotta M, Malatesta S, Chella A, Barassi F, Mucilli F, Camplese P, D'Antuono T, Sacco R and Buttitta F (2008) Frequent mutations in the neurotrophic tyrosine receptor kinase gene family in large cell

neuroendocrine carcinoma of the lung. Hum Mutat 29:609-616. at ASPET Journals on September 30, 2021 Moschetta M, Cesca M, Pretto F and Giavazzi R (2010) Angiogenesis inhibitors: implications for combination with conventional therapies. Curr Pharm Des 16:3921-3931. Olaso E, Lin HC, Wang LH and Friedman SL (2011) Impaired dermal wound healing in discoidin domain receptor 2-deficient mice associated with defective extracellular matrix remodeling. Fibrogenesis Tissue Repair 4:5. Quentmeier H, Reinhardt J, Zaborski M and Drexler HG (2003) FLT3 mutations in acute myeloid leukemia cell lines. Leukemia 17:120-124. Reck M, Kaiser R, Mellemgaard A, Douillard JY, Orlov S, Krzakowski M, von Pawel J, Gottfried M, Bondarenko I, Liao M, Gann CN, Barrueco J, Gaschler-Markefski B, Novello S and Group LU-LS (2014) Docetaxel plus nintedanib versus docetaxel plus placebo in patients with previously treated non-small-cell lung cancer (LUME-Lung 1): a phase 3, double-blind, randomised controlled trial. Lancet Oncol 15:143-155. Reinmuth N, Heigener D and Reck M (2015) Novel angiogenesis inhibitors in nonsmall cell lung cancer. Curr Opin Oncol 27:79-86. Reva B, Antipin Y and Sander C (2011) Predicting the functional impact of protein mutations: application to cancer genomics. Nucleic Acids Res 39:e118. Rocco G, Morabito A, Leone A, Muto P, Fiore F and Budillon A (2016) Management of non-small cell lung cancer in the era of personalized medicine. Int J Biochem Cell Biol 78:173-179. Stankov K, Popovic S and Mikov M (2014) C-KIT signaling in cancer treatment. Curr Pharm Des 20:2849-2880. Stukalin I, Alimohamed N and Heng DY (2016) Contemporary Treatment of Metastatic Renal Cell Carcinoma. Oncol Rev 10:295. Suzuki M, Makinoshima H, Matsumoto S, Suzuki A, Mimaki S, Matsushima K, Yoh K, Goto K, Suzuki Y, Ishii G, Ochiai A, Tsuta K, Shibata T, Kohno T, Esumi H and Tsuchihara K (2013) Identification of a lung adenocarcinoma cell line with CCDC6-RET fusion gene and the effect of RET inhibitors in vitro and in vivo. Cancer Sci 104:896-903. JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

27 JPET #244129

Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, Salzberg SL, Wold BJ and Pachter L (2010) Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 28:511-515. Weiss J, Sos ML, Seidel D, Peifer M, Zander T, Heuckmann JM, Ullrich RT, Menon R, Maier S, Soltermann A, Moch H, Wagener P, Fischer F, Heynck S, Koker M, Schottle J, Leenders F, Gabler F, Dabow I, Querings S, Heukamp LC, Balke-Want H, Ansen S, Rauh D, Baessmann I, Altmuller J, Wainer Z, Conron M, Wright G, Russell P, Solomon B, Brambilla E, Brambilla C, Lorimier P, Sollberg S, Brustugun OT, Engel-Riedel W, Ludwig C, Petersen I, Sanger J, Clement J, Groen H, Timens W, Sietsma H, Thunnissen E, Smit E, Heideman D, Cappuzzo F, Ligorio C, Damiani S, Hallek M, Beroukhim R, Pao W, Klebl B, Baumann M, Buettner R, Ernestus K, Stoelben E, Wolf J, Nurnberg P, Perner S and Thomas RK (2010) Frequent and focal FGFR1 amplification associates with therapeutically tractable FGFR1 dependency in squamous cell lung cancer. Sci Transl Med 2:62ra93. Winterhoff B and Konecny GE (2016) Targeting fibroblast growth factor pathways in endometrial

cancer. Curr Probl Cancer. Downloaded from Wood JM, Bold G, Buchdunger E, Cozens R, Ferrari S, Frei J, Hofmann F, Mestan J, Mett H, O'Reilly T, Persohn E, Rosel J, Schnell C, Stover D, Theuer A, Towbin H, Wenger F, Woods-Cook K, Menrad A, Siemeister G, Schirner M, Thierauch KH, Schneider MR, Drevs J, Martiny-Baron G and Totzke F (2000) PTK787/ZK 222584, a novel and potent inhibitor of vascular endothelial

growth factor receptor tyrosine kinases, impairs vascular endothelial growth factor-induced jpet.aspetjournals.org responses and tumor growth after oral administration. Cancer Res 60:2178-2189.

at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

28 JPET #244129

Footnotes:

The authors declare no potential conflicts of interest. FH, UT-G, AB, SL, DG, TV, PG-C, CH and NK are employees of Boehringer Ingelheim.

Reprint requests: Frank Hilberg, Boehringer Ingelheim RCV GmbH Co KG, Dr. Boehringer-

Gasse 5-11, A-1121 Vienna, Austria, Phone: ++43-1-80105-2351, Fax: ++43-1-80105-2366 e-mail: [email protected]

Downloaded from jpet.aspetjournals.org at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

29 JPET #244129

Figure Legends

Figure 1. Inhibition of Proliferation and TRKA and downstream signaling of TRKA fusion cell lines using nintedanib and entrectinib. Human TRKA fusion cell lines CUTO3.29 (A) and

KM12 (B) were treated with the indicated dose range of ninitedanib (blue) or entrectinib (red) and assayed for cell proliferation 72 hours after treatment using an MTS assay (n = 3, error bars represent ± S.E.M.). The half-maximal inhibitory concentration (IC50) was calculated for

each cell line as follow: CUTO3.29 treated with nintedanib, 954.9 ± 7.2 nmol/l or with Downloaded from entrectinib, 117.48 ± 5.6 nmol/l; KM12 treated with nintedanib, 557.1 ± 5.6 nmol/l, and with entrectinib, 2.28 ± 0.3 nmol/l. CUTO3.29 (C) and KM12 (D) were treated with the indicated increasing doses of either nintedanib or entrectinib and protein lysates were collected 2 jpet.aspetjournals.org hours later for western blot analysis of phospho-TRKA and downstream signaling of phospho-ERK1/2, phospho-AKT and phospho-STAT3. Western blot images are representative of 3 independent experiments. at ASPET Journals on September 30, 2021

Figure 2. Mutational analysis of nintedanib sensitive tumor cell lines. A: Response of Ricerca

240-OncoPanel™ cell lines to nintedanib. Sensitive cell lines (GI50 ≤ 500 nmol/l) are marked with *. The color code legend highlights the tumor types represented by the sensitive cell lines. B: Venn diagram showing the 16 genes and their respective alterations in the sensitive cell lines highlighted in panel A.

Figure 3. Nintedanib inhibits ligand-dependent phosphorylation of MAPK and Akt in NCI-

H1703 and Kato III tumor cells. A: NCI-H1703 Western blot analysis after exposure to either nintedanib or imatinib following stimulation with PDGF BB. Strong concentration-dependent reduction of phosphorylated MAPK and Akt levels by nintedanib as compared to imatinib. B:

NCI-H1703 Western blot analysis after 72h of either FGFR1 or PDGFRA siRNA treatment and 2h of exposure to nintedanib shows concentration-dependent reduction of phosphorylated MAPK and Akt levels. Cleaved PARP and cleaved Caspase-3 serve as JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

30 JPET #244129 indicators for apoptosis. Nintedanib concentrations are shown in μmol/L. C: Kato III Western blot analysis after exposure to either nintedanib or following stimulation with bFGF. Slightly stronger concentration-dependent reduction of phosphorylated MAPK and Akt levels by

PD173074 as compared to nintedanib. Western blot images are representative of 3 independent experiments. D: NCI-H1703 Western blot analysis after exposure to either nintedanib, sunitinib or sorafenib following stimulation with PDGF BB. Concentration- dependent reduction of phosphorylated MAPK, Akt and phosphor PDGFR levels by nintedanib and sunitinib as compared to sorafenib. Downloaded from

Figure 4. Single agent nintedanib induces tumor shrinkage of subcutaneous NCI-H1703 tumors. A) Median tumor volume over time, B) Single tumor volumes + median at day 29, jpet.aspetjournals.org nintedanib, 100 mg/kg qd (open triangles), vatalanib, 100 mg/kg qd (stars), imatinib, 75 mg/kg qd (filled triangles) and imatinib, 75 mg/kg bid (filled diamonds) (n=7). Nintedanib in combination with the BET inhibitor BI894999 shows excellent anti-tumor efficacy with 7 of 8 at ASPET Journals on September 30, 2021 animals surviving ≥ 100 days. C) Median tumor volume over time. (Nintedanib treated animals euthanized prematurely because of tumor necrosis.) D) Single tumor volumes + median at day 33. E) Survival probability (n=8 for treatment, n=10 for vehicle control).

Xenografts were established s.c. in athymic mice and allowed to reach a volume of ~100 mm³ before treatment JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

31 JPET #244129

Table 1A, Potent inhibition of triple angiokinase targets by nintedanib and competitor compounds

Kinase § VEGFR PDGFR FGFR

IC50 [nmol/l] -1 -2 -3 alpha beta -1 -2 -3 -4 nintedanib 99 3 4 18 28 41 47 96 421 sunitinib 629 96 63 71 38 531 184 559 2670 sorafenib 765 100 274 174 1110 >10000 994 >10000 >10000 vandetanib 208 12 74 187 3190 804 149 970 9560 Downloaded from pazopanib 171 19 94 173 62 321 353 904 1660 cediranib 64 4 18 296 496 39 20 63 622

Performed at Invitrogen, [ATP] @ Km app jpet.aspetjournals.org § additional kinases tested: ABL1, ACVR1B (ALK4), AKT2 (PKB beta), AMPK A1/B1/G1, AURKA (Aurora A), CAMK1D (CaMKI delta), CDK2/cyclin A, CHEK1 (CHK1), CSNK1A1 (CK1 alpha 1), CSNK2A1 (CK2 alpha 1), EGFR (ErbB1), EPHB2, FGFR1, FRAP1 (mTOR),

GSK3B (GSK3 beta), IGF1R, LCK, MAP2K1 (MEK1), MAP3K8 (COT), MAPK14 (p38 alpha), at ASPET Journals on September 30, 2021 MAPKAPK2, MYLK2 (skMLCK), NEK2, PAK4, PDK1, PRKACA (PKA), RAF1 (cRAF) Y340D Y341D, ROCK2, SRPK2, STK3 (MST2), TBK1

Table 1B, Nintedanib potently inhibits the proliferation of cell lines carrying alterations in PDGFRA or FGFR2

NCI-1703 KatoIII AN3CA MFM-223 EC [nmol/l] 50 PDGFRA ampl FGFR2 ampl FGFR2 mut FGFR2 ampl nintedanib 10 ± 3 176 ± 40 152 ± 22 108

PD173074 n.a. 30 ± 5 n.a. n.a. imatinib 109 ± 65 n.a. n.a. n.a. sunitinib 39 ± 10 624 ± 85 n.a. n.a. sorafenib 258 ± 25 383 ± 24 n.a. n.a. Data were determined using an alamarBlue® Cell Viability Assay Protocol from the Cell Proliferation Assay Protocols section of Life Technologies, drug exposure for 72 h JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version.

32 JPET #244129

Table 2: Extended in vitro kinase inhibition profile of nintedanib

Kinase IC50 (nmol/L)

ABL1 12 ± 5

BLK 42 ± 9

BTK 34 ± 14

CSF1R 5 ± 2

DDR1 17 ± 7 Downloaded from DDR2 16 ± 4

FYN 74 ± 24

JAK3 67 ± 27 jpet.aspetjournals.org

KIT 6 ± 3

MAP3K3 (MEKK3) 58 ± 25

MAP3K7 46 ± 31 at ASPET Journals on September 30, 2021

MELK 3 ± 2

MST4 84 ± 9

NTRK1 (TRKA) 30 ± 8

NTRK3 (TRKC) 48 ± 25

NUAK1 50 ± 8

RET 2 ± 1

SIK2 11 ± 4

STK24 61 ± 12

TGFBR1 77 ± 20

YES1 14 ± 4

Data were determined using the SelectScreen® Biochemical Kinase Profiling Service and performed at ThermoFisher/life technologies. Measurements were performed in triplicates.

[ATP] @ Km app JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version. Downloaded from

33 JPET #244129

jpet.aspetjournals.org at ASPET Journals on September 30, 2021

JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version. Downloaded from

34 JPET #244129

Hilberg et al., Figure 2 jpet.aspetjournals.org

B at ASPET Journals on September 30, 2021 overexpression

FLT4 CSF1R, MAP3K3, NUAK1 NTRK1 PDGFRB

KDR KIT BTK,DDR2, FGFR1, FLT1, FLT3-ITD, FYN, FGFR2 NTRK3, PDGFRA

amplification mutation JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version. Downloaded from

35 JPET #244129

Hilberg et al., Figure 3A jpet.aspetjournals.org

A at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version. Downloaded from

36 JPET #244129

jpet.aspetjournals.org Hilberg et al., Figure 3B

B at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version. Downloaded from

37 JPET #244129

jpet.aspetjournals.org

Hilberg et al., Figure 3C

C at ASPET Journals on September 30, 2021 JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version. Downloaded from

38 JPET #244129

jpet.aspetjournals.org at ASPET Journals on September 30, 2021

JPET Fast Forward. Published on December 20, 2017 as DOI: 10.1124/jpet.117.244129 This article has not been copyedited and formatted. The final version may differ from this version. Downloaded from

39 JPET #244129 Hilberg et al., Figure 4 d a y 2 9

A 1400 p = 0 .0 0 7 0 B jpet.aspetjournals.org 1200 p = 0 .0 0 4 6

] 2 5 0 0 3 1000 p = 0 .0 0 4 6

] 2 0 0 0 800 3

control m

m

[

600 nintedanib e 1 5 0 0

at ASPET Journals on September 30, 2021 m

vatalanib u

l o

400 imatinib qd V

1 0 0 0 r

imatinib bid o

200 m

u T

Median Tumor Volume [mm Volume Tumor Median 5 0 0 0 0 5 10 15 20 25 30 Days 0 l o ib ib d id tr n n q b n a la b o d a i ib c e t in n 1600 t a t ti in v a a n m C i im ] 3 1400 1200 control d a y 3 3 1000 D BI 894999 800 p < 0 .0 0 0 2 600 nintedanib 2 0 0 0 p = 0 .0 0 1 9

400 BI 894999 + ]

200 nintedanib 3 1 5 0 0 m

Median Tumor Volume Tumor [mm Median Volume

m

[

0 e m

0 20 40 60 80 100 u l 1 0 0 0

Day o

V

1 0 0 r o

m E u 5 0 0

8 0 T

] %

[ 6 0

l 0

a c o n tr o l

v b i 9 b i 9 i v n n

r B I 8 9 4 9 9 9 9 a a 4 0 4 d u 9 d e te S 8 t n in te d a n ib I in n B n n + 9 2 0 B I 8 9 4 9 9 9 + n in te d a n ib 9 9 4 9 8 I B 0 0 2 0 4 0 6 0 8 0 1 0 0 D a y s 1 JPET #244129

The Journal of Pharmacology and Experimental Therapeutics

The triple angiokinase inhibitor nintedanib directly inhibits tumor cell growth and induces tumor shrinkage via blocking oncogenic receptor tyrosine kinases

Frank Hilberg, Ulrike Tontsch-Grunt, Anke Baum, Anh T. Le, Robert C. Doebele, Simone

Lieb, Davide Gianni, Tilman Voss, Pilar Garin-Chesa, Christian Haslinger and Norbert Kraut

Boehringer Ingelheim RCV GmbH Co KG, Dr. Boehringer-Gasse 5-11, A-1121 Vienna,

Austria (F.H., U.T-G, A.B., S.L., D.G., P.G-C, C.H., N.K.)

University of Colorado, School of Medicine, Division of Medical Oncology, 12801 E. 17th

Ave., MS 8117, Aurora, CO 80045 (A.T.L., R.C.D.)

AstraZeneca - Innovative Medicines and Early Development | Discovery Sciences, 310 Unit

Darwin Building - Cambridge Science Park - Milton Rd - CB4 0WG (D.G.)

JPET #244129

Hilberg et al., Supplementary Tables:

Table S1a Sensitive cell lines: sensitivity, tumor type, altered genes Cell Line GI50 (nmol/l) tumortype histology_type histology_subtype altered gene A-204 21,79 rhabdomyosarcoma rhabdomyosarcoma FLT4, NTRK1 A-427 311,8 NSCLC adenocarcinoma 418 mutated genes BV-173 405,3 hematopoietic/leukemia myeloid leukemia CML, chronic myelogenous leukemia FLT4, KIT, PDGFRB CGTH-W-1 8,883 thyroid carcinoma carcinoma follicular KDR, KIT CML-T1 155,2 hematopoietic/leukemia myeloid leukemia CML, chronic myelogenous leukemia BTK, FLT1, FYN, NTRK1 EM-2 174,8 hematopoietic/leukemia myeloid leukemia CML, chronic myelogenous leukemia MAP3K3 G-402 7,712 renal cancer other leiomyoblastoma 311 mutated genes

KATO III 209,2 gastric carcinoma carcinoma gastric DDR2, FGFR2, FLT1 MEG-01 420,5 hematopoietic/leukemia myeloid leukemia CML, chronic myelogenous leukemia FGFR1, KDR, KIT, NTRK1 MV-4-11 52,88 hematopoietic/leukemia lymphoid leukemia AML, biphenotypic B myelomonocytic FLT3-ITD, CSF1R, leukemia NUAK1 SJCRH30 219,6 rhabdomyosarcoma rhabdomyosarcoma PDGFRA, PDGFRB SW579 21,98 thyroid carcinoma squamous cell carcinoma KIT

JPET #244129

Table S1b Sensitive cell lines and their respective indications tumortype sensitive insensitive total gastric carcinoma 1 5 6 NSCLC 1 14 15 renal cancer other 1 2 3 rhabdomyosarcoma 2 2 4 thyroid carcinoma 2 2 4 hematopoietic/leukemia 5 12 17 12

JPET #244129

Table S2 Distribution of alterations in sensitive cell lines (m: mutation; a: amplification; e: overexpression) cell line BTK CSF1R DDR2 FGFR1 FGFR2 FLT1 FLT3 FLT4 FYN KDR KIT MAP3K3 NTRK1 NTRK3 NUAK1 PDGFRA PDGFRB A-204 m m BV-173 e m e CGTH-W-1 a a CML-T1 m m m m EM-2 e KATO III m a m MEG-01 m e e e MV-4-11 e m e SJCRH30 m m SW579 a

JPET #244129

Table S3 celllinename TumorType EC50 [nM] 22Rv1 prostate carcinoma 2727 5637 bladder carcinoma 5594 639-V bladder carcinoma 1886 647-V bladder carcinoma 3775 769-P renal carcinoma 5475 786-O renal carcinoma 3848 A101D melanoma 8008 A-172 astrocytoma/glioblastoma 2742 A-204 rhabdomyosarcoma 10,02 A-375 melanoma 3429 A-427 NSCLC 409,1 A-431 skin/SCC 454 A-498 renal carcinoma 6413 A549 NSCLC 4230 A-673 sarcoma/soft tissue 4123 A7 melanoma 4291 ACHN renal carcinoma 5856 AGS gastric carcinoma 2731 AN3 CA uterus carcinoma 1529 ARH-77 hematopoietic/leukemia 4489 AsPC-1 pancreas carcinoma 4449 AU565 breast carcinoma 3872 BC-1 hematopoietic/lymphoma 3064 BE(2)-C neuroblastoma 622,2 BeWo placenta carcinoma 4249 BFTC-905 bladder carcinoma 3370 BHT-101 thyroid carcinoma 3895 BM-1604 prostate carcinoma 10000 JPET #244129

BPH-1 prostate benign hyperplasia 1497 BT-20 breast carcinoma 6384 BT-474 breast carcinoma 3173 BT-483 breast carcinoma 3590 BT-549 breast carcinoma 3544 BV-173 hematopoietic/leukemia 429,8 BxPC-3 pancreas carcinoma 4234 C32 melanoma 7793 C32TG melanoma 4033 C-33 A cervix carcinoma 2956 C-4 I cervix carcinoma 3782 C-4 II cervix carcinoma 4226 Caki-1 renal carcinoma 3847 Caki-2 renal carcinoma 3895 CAL-27 HNSCC 4049 CAL-62 thyroid carcinoma 3616 Calu-1 NSCLC 5363 Calu-6 NSCLC 3329 CAMA-1 breast carcinoma 4163 Caov-3 ovarian carcinoma 9140 Capan-1 pancreas carcinoma 5267 Capan-2 pancreas carcinoma 4573 CCF-STTG1 astrocytoma/glioblastoma 9512 CCRF-CEM hematopoietic/leukemia 3645 CEM/C1 hematopoietic/leukemia 3076 CFPAC-1 pancreas carcinoma 4936 CGTH-W-1 thyroid carcinoma 8,969 ChaGo-K-1 NSCLC 8175 CHL-1 melanoma 1122 CHP-212 neuroblastoma 1039 JPET #244129

CML-T1 hematopoietic/leukemia 186,2 COLO 201 colon carcinoma 8177 COLO 205 colon carcinoma 4257 COLO 320DM colon carcinoma 5102 COLO 320HSR colon carcinoma 8096 COLO 829 melanoma 8817 COR-L105 NSCLC 1657 COR-L23 NSCLC 10000 CRO-AP2 hematopoietic/lymphoma 1353 D283 Med medulloblastoma 2370 Daoy medulloblastoma 4256 Daudi hematopoietic/lymphoma 2288 DB hematopoietic/lymphoma 3384 DBTRG-05MG astrocytoma/glioblastoma 10000 Detroit 562 HNSCC 3775 DK-MG astrocytoma/glioblastoma 7927 DLD-1 colon carcinoma 6002 DMS 114 SCLC 1743 DMS 273 SCLC 3777 DMS 53 SCLC 4126 DOHH-2 hematopoietic/lymphoma 1081 DoTc2 4510 cervix carcinoma 5299 DU 145 prostate carcinoma 5052 EB-3 hematopoietic/lymphoma 2444 EFM-19 breast carcinoma 3344 EM-2 hematopoietic/leukemia 182,5 ES-2 ovarian carcinoma 3225 FaDu HNSCC 5048 G-401 renal cancer other 1269 G-402 renal cancer other 12,67 JPET #244129

H4 astrocytoma/glioblastoma 7826 HCT 116 colon carcinoma 1828 HCT-15 colon carcinoma 3911 HCT-8 colon carcinoma 1937 HEC-1-A uterus carcinoma 4791 HEL 92.1.7 hematopoietic/leukemia 9434 HeLa S3 cervix carcinoma 4509 Hep G2 liver carcinoma 2133 HL-60/MX1 hematopoietic/leukemia 2120 HLE liver carcinoma 1143 HLF liver carcinoma 1284 HMCB melanoma 657,8 HOS bone sarcoma 1623 HPAF-II pancreas carcinoma 3583 Hs 578T breast carcinoma 2274 Hs 695T melanoma 10000 Hs 746T gastric carcinoma 8615 Hs 766T pancreas carcinoma 4519 HT hematopoietic/lymphoma 10000 HT-1080 sarcoma/soft tissue 3468 HT-1197 bladder carcinoma 6441 HT-1376 bladder carcinoma 5296 HT-29 colon carcinoma 2262 HT-3 cervix carcinoma 2927 HuCCT1 liver carcinoma 3363 HUH-6 Clone 5 liver carcinoma 1049 HuP-T4 pancreas carcinoma 4448 IMR-32 neuroblastoma 2558 J82 bladder carcinoma 5762 JAR placenta carcinoma 3672 JPET #244129

JEG-3 placenta carcinoma 3729 J.RT3-T3.5 hematopoietic/leukemia 3716 Jurkat, Clone E6-1 hematopoietic/leukemia 4023 K-562 hematopoietic/leukemia 782,6 KATO III gastric carcinoma 204,6 KHOS-240S bone sarcoma 1391 KLE uterus carcinoma 10000 KPL-1 breast carcinoma 1331 L-428 hematopoietic/lymphoma 10000 LNCap.FGC prostate carcinoma 1398 LS1034 colon carcinoma 4265 LS 174T colon carcinoma 2743 Malme-3M melanoma 4699 MCF7 breast carcinoma 2247 MC-IXC neuroblastoma 3420 MDA-MB-231 breast carcinoma 5668 MDA-MB-436 breast carcinoma 10000 MDA-MB-453 breast carcinoma 5119 MDA-MB-468 breast carcinoma 3618 MEG-01 hematopoietic/leukemia 434,6 MES-SA sarcoma/soft tissue 3417 MeWo melanoma 3932 MG-63 bone sarcoma 1520 MHH-PREB-1 hematopoietic/lymphoma 4923 MIA PaCa-2 pancreas carcinoma 1750 MOLT-16 hematopoietic/leukemia 3518 MOLT-3 hematopoietic/leukemia 10000 MT-3 breast carcinoma 2375 MV-4-11 hematopoietic/leukemia 57,56 NALM-6 hematopoietic/leukemia 1898 JPET #244129

NCI-H292 NSCLC 1184 NCI-H295R adrenal gland carcinoma 10000 NCI-H441 NSCLC 10000 NCI-H446 SCLC 3023 NCI-H460 NSCLC 4793 NCI-H508 colon carcinoma 4754 NCI-H520 NSCLC 1303 NCI-H596 NSCLC 10000 NCI-H661 NSCLC 3206 NCI-H69 SCLC 3230 NCI-H747 colon carcinoma 3545 OCUG-1 gallbladder carcinoma 10000 OE19 esophagus carcinoma 3271 OE21 esophagus carcinoma 3661 OVCAR-3 ovarian carcinoma 10000 PANC-1 pancreas carcinoma 3014 PC-3 prostate carcinoma 5636 Raji hematopoietic/lymphoma 4860 Ramos (RA1) hematopoietic/lymphoma 4731 RD rhabdomyosarcoma 1443 RKO colon carcinoma 3640 RKO-AS45-1 colon carcinoma 3226 RKO-E6 colon carcinoma 3062 RL hematopoietic/lymphoma 2189 RL95-2 uterus carcinoma 8985 RPMI 6666 hematopoietic/lymphoma 2312 RPMI-7951 melanoma 3411 RPMI 8226 hematopoietic/myeloma 4407 SCaBER bladder carcinoma 4313 SCC-25 HNSCC 2397 JPET #244129

SCC-4 HNSCC 3814 SCC-9 HNSCC 7757 SH-4 melanoma 2683 SHP-77 SCLC 3219 SiHa cervix carcinoma 5374 SJCRH30 rhabdomyosarcoma 356,8 SJSA-1 bone sarcoma 1721 SK-BR-3 breast carcinoma 3365 SK-LMS-1 sarcoma/soft tissue 3711 SK-MEL-1 melanoma 9207 SK-MEL-28 melanoma 5050 SK-MEL-3 melanoma 3589 SK-MES-1 NSCLC 3515 SK-N-AS neuroblastoma 4309 SK-N-DZ neuroblastoma 2478 SK-NEP-1 renal cancer other 9579 SK-N-FI neuroblastoma 3122 SKO-007 hematopoietic/myeloma 4041 SK-OV-3 ovarian carcinoma 8807 SK-UT-1 sarcoma/soft tissue 5058 SNB-19 astrocytoma/glioblastoma 3974 SNU-1 gastric carcinoma 1092 SNU-16 gastric carcinoma 565,2 SNU-423 liver carcinoma 3450 SNU-5 gastric carcinoma 10000 SR hematopoietic/lymphoma 975,8 ST486 hematopoietic/lymphoma 3600 SU.86.86 pancreas carcinoma 7573 SW 1088 astrocytoma/glioblastoma 3754 SW1116 colon carcinoma 10000 JPET #244129

SW-13 adrenal gland carcinoma 3403 SW 1353 bone sarcoma 1683 SW1417 colon carcinoma 4910 SW1463 colon carcinoma 5000 SW 1783 astrocytoma/glioblastoma 10000 SW403 colon carcinoma 6510 SW48 colon carcinoma 2028 SW480 colon carcinoma 10000 SW579 thyroid carcinoma 38,81 SW620 colon carcinoma 4589 SW 684 sarcoma/soft tissue 6832 SW837 colon carcinoma 3587 SW 872 sarcoma/soft tissue 3898 SW 900 NSCLC 926,8 SW948 colon carcinoma 4082 SW 954 vulva carcinoma 2655 SW962 vulva carcinoma 1688 SW 982 sarcoma/soft tissue 8450 T24 bladder carcinoma 4738 T-47D breast carcinoma 8031 T84 colon carcinoma 9000 T98G astrocytoma/glioblastoma 962,5 TCCSUP bladder carcinoma 2225 TE 381.T rhabdomyosarcoma 1023 THP-1 hematopoietic/leukemia 7485 U-138 MG astrocytoma/glioblastoma 3673 U266B1 hematopoietic/myeloma 1740 U-2 OS bone sarcoma 1394 U-87 MG astrocytoma/glioblastoma 5485 UM-UC-3 bladder carcinoma 3303 JPET #244129

WI-38 normal 1730 WiDr colon carcinoma 1549 Y79 retinoblastoma 5808 YAPC pancreas carcinoma 4421

JPET #244129

Hilberg et al.: Supplementary Figures:

Supplementary Figure S1 Cross cancer alteration summary of all 16 kinases with alterations in the screened cell lines, AML: Acute Myeloid Leukemia, ALL: Acute Lymphoblastic Leukemia, RMS: Rhabdomyosarcoma, STAC: Stomach Adenocarcinoma, THR: Thyroid Cancer

JPET #244129

gastric carcinoma acute lymphoblastic leukemia acute myeloid leukemia thyroid cancer

Supplementary Figure S2 Expression of selected genes in human tumor types corresponding to the respective tumor types of the sensitive cell lines. The numbers indicate numbers of samples with over expression versus all samples in the respective tumor type. Overexpression threshold is defined as : mean(normal tissue) + 3 * stddev(normal tissue)

JPET #244129 A total copynumber: 5.5

0 0 2 4 6 8 10

B

total copynumber: 5.5

absolute copynumber absolute

0 0 2 4 6 10 8

C total copynumber: 14 0 0 2 4 6 8 10 chromosomal position

Supplementary Figure S3 A: focal amplification of PDGFRA in NCI-H1703 cells. B: focal amplification of FGFR1 in NCI-H1703 cells. C: focal amplification of FGFR2 in tetraploid KATO III cells. The red line marks the position of the respective gene. JPET #244129

Supplementary Figure S4 Visualization and measurement of cell growth in AML cell lines: A) MOLM13, B) MV-4-11-B and C) THP1 by IncuCyte ZOOM® live cell imaging. Filled circles, DMSO @ 0.1%; filled upward triangles, Nintedanib @ 33 nM; filled squares, BI 894999 @ 10 nM; filled downward triangles, combination of BIBF1120 @ 33 nM and BI 894999 @ 10 nM. JPET #244129

Hilberg et al.: Supplementary Method information:

20120701 SSBK Customer Protocol and Assay Conditions.pdf

20120701 SSBK-Adapta Customer Protocol and Assay Conditions.pdf

20120701 SSBK-LanthaScreen Binding Customer Protocol and Assay Conditions.pdf