ROS1 POSITIVE CANCER TREATMENT

FIELD OF THE INVENTION The invention relates to cancer treatment using a ROS1 inhibitor.

BACKGROUND OF THE INVENTION

Receptor tyrosine kinases (RTKs) are a large family of cell surface receptors that sense growth factors and hormones and regulate a variety of cell behaviors, such as cell proliferation and survival. Unregulated and constitutive RTK activation through chromosomal rearrangements, point mutations, and gene amplification has been shown to be responsible for the initiation and progression of many cancers and other diseases. In turn, defective RTK signaling has been linked to degenerative diseases. The ros oncogene encodes an orphan RTK related to anaplastic lymphoma kinase (ALK), along with members of the insulin-receptor family [1 ]. First discovered as the oncogene product of an avian sarcoma RNA tumor virus [2, 3], c-ROS proto-oncogene 1 receptor tyrosine kinase (ROS1 ) presents elevated expression levels in 20-30% of patients with non-small cell lung cancer (NSCLC), which accounts for 80% of cases of lung cancer [4], and in 13% of patients with lung adenocarcinoma [5]. In -2% of NSCLC cases, ROS1 signaling is constitutively activated by interchromosomal translocation or intrachromosomal deletion that results in ROS1 fusion genes. Several ROS1 kinase domain fusion proteins have been identified, including the Fused in Glioblastoma-ROSI (FIG-ROS), first in a human glioblastoma cell line [6] and later in patients with NSCLC [7], cholangiocarcinoma [8], and serous ovarian carcinoma [9]. The SLC34A2-ROS1 (SLC-ROS) fusion is present in a subset of patients with NSCLC [10, 1 1 ] and gastric cancer [12]. Other ROS1 fusions include CD74-ROS1 , EZR- ROS1 , LRIG3-ROS1 , SDC4-ROS1 , TPM3-ROS1 , among others [13]. Notably, patients with ROS1 rearrangements are significantly younger and more likely to be never-smokers[10] .

Traditionally, depending on the type of tumor, therapeutic approaches include different combinations of surgery, radiation therapy, and chemotherapy. However, alternative therapies using RTKs as targets started to be introduced in the beginning of this century. In the case of lung cancer, such approaches have led to discovery of a number of these targeted therapies with specific inhibitor drugs such as erlotinib and gefitinib for the epidermal growth factor receptor (EGFR) [14, 15]. However, in lung adenocarcinomas with the presence of ALK and/or ROS1 rearrangements these treatments are ineffective. This problem has been partially overcome due to the treatment of ALK rearrangements-positive cancers with crizotinib (Xalkori®, Pfizer) [16]. A recent phase l/l I clinical trial designed to evaluate the safety and efficacy of crizotinib in patients with ROS1 fusion-positive lung cancer demonstrated a high response rate towards improvement of patients with ROS1 -rearrangements [17, 18].

Recent evidence showed that a subset of patients with crizotinib-treated ROS1 fusion-positive tumor acquired a ROS1 kinase domain mutation (ROS1 ^G2032R ) that confers crizotinib resistance [19, 20]. This situation forces the identification of small molecules to prevent the manifestation of ROS1 mutations or to inhibit the new resistant variant.

To identify additional and potentially more effective ROS1 and ROS1 mutants inhibitors, researchers employed unbiased, high-throughput kinase inhibitor screening assay and discovered that PF-06463922 (originally an ALK inhibitor) [21 , 22], cabozantinib (originally a cMET/RET/VEGFR inhibitor) [23] and foretinib (originally a cMETA/EGFR inhibitor) [24] are potent inhibitors of ROS1 and ROS1 ^G2032R These drugs selectively suppress receptor activity when tested in cellular-based assays. PF- 06463922 and cabozantinib are currently undergoing phase l/ll clinical trial investigation (ClinicalTrials.gov Id: NCT01970865 and NCT01639508, respectively) against ALK/ROS1 positive NSCLC. Furthermore, cabozantinib is currently clinically available to treat medullary thyroid cancer against cMET, VEGFR2 and RET (ClinicalTrials.gov Id: NCT016831 10). However, none of them has been currently approved for the use in ROS1 -fusion positive tumors. Therefore, there is still an urgent need for the discovery of new agents to target ROS1 fusion-positive cancers, especially for those that are effective against ROS1 mutants.

Among the ROS1 positive tumors are malignant epithelial tumors, such as in NSCLC, gastric cancer, serous ovarian carcinoma or cholangiocarcinoma.

Rothenstein et al. [25] summarize the management of treatment related adverse events that can arise with ALK inhibitors such as crizotinib and second-generation ALK inhibitors. Although those inhibitors are generally well tolerated, they have a unique side effect profile that differs from that of traditionally cytotoxic therapy and treatment will be often for long periods of time. Katayama et al. [23] describe a model of acquired resistance to ROS1 inhibitors in NSCLC with ROS1 rearrangement and identified cabozantinib as a therapeutic strategy to overcome the resistance by high throughput drug screening with small molecular inhibitors and anticancer drugs used in clinical practice or being currently tested in clinical trials.

Molecular docking studies using EML4-ALK translocation mediated fusion protein were disclosed by Ramshankar et al. [26]. Tivozanib and lapatinib were found to potentially inhibit the EML4-ALK, suggesting further studies whether tivozanib and lapatinib indeed bind the ALK region.

Tivozanib (AV-951 , KRN951 ) is known as a potent, selective, long half-life inhibitor of all three vascular endothelial growth factor (VEGF) receptors that is designed to optimize VEGF blockade while minimizing off -target toxicities [27]. It has previously been shown that tivozanib has the activity of reversing multidrug resistance mediated by ABCB1 (p-glycoprotein) and ABCG2 transporters (BCRP), two common mechanism of resistance against small molecules [28]. However, data found in online repositories indicate that tivozanib is not efficiently inhibiting ALK.

WO2013/158859A1 discloses treatment of NSCLC, which can express ALK and ROS1 in a mutually exclusive way. Crizotinib or TAE-684 inhibiting both, ROS1 activity and ALK activity, is suggested for treating NSCLC.

AU2015100 840A4 discloses an oncogenic ROS1 kinase inhibitor that exhibits anti-cancer activity in an NSCLC cell with ROS1 fusion gene.

US2014/243332A1 discloses treatment of cancers characterized by aberrant ROS1 activity by administering an effective amount of foretinib. SUMMARY OF THE INVENTION

It is the object to provide an improved therapy of ROS1 positive cancer patients. It is the specific object to provide a molecularly targeted therapy directed to inhibit ROS1 positive cancer, in particular both, ROS1 and ROS1 ^G2032R mutant. The object is solved by the subject matter of the claims.

The invention provides for a ROS1 inhibitor for use in the treatment of a subject suffering from ROS1 positive cancer, wherein the inhibitor is tivozanib.

Specifically, tivozanib is dosed at 0.5-2.0 mg daily, preferably at 1 .5 mg daily. Specifically, tivozanib is administered orally one dose per day for three weeks, optionally followed by one week off treatment (i.e., without tivozanib administration).

Specifically, tivozanib is 1 -{2-Chloro-4-[(6,7-dimethoxyquinolin-4-yl)oxy]phenyl}- 3-(5-methylisoxazol-3-yl)urea or 1 -{2-Chloro-4-[(6,7-dimethoxyquinolin-4- yl )oxy]phenyl}-3-(5-methyl isoxazol-3-yl ) urea monohydrochloric acid salt monohydrate.

Specifically, tivozanib is administered in combination with another inhibitor targeting any of ROS , c-MET, VEGF or RET. Said another inhibitor is preferably an antagonist molecule specifically binding to any of ROS1 , c-MET, VEGF or RET, preferably wherein said antagonist molecule is any of a small molecule, polypeptide, peptide, nucleic acid, or oligonucleotide, specifically a chemotherapeutic agent and/or a therapeutic antibody.

According to a specific embodiment, the chemotherapeutic agent is selected from mitotic inhibitors, alkylating agents, anti-metabolites, proteasome inhibitor, intercalating antibiotics, growth factor inhibitors, cell cycle inhibitors, enzymes, topoisomerase inhibitors, biological response modifiers, anti-hormones, angiogenesis inhibitors, and anti-androgens.

According to a specific embodiment, the therapeutic antibody is specifically recognizing a human target selected from the group consisting of vascular endothelial growth factor (VEGF) or its receptor (VEGFR), epidermal growth factor receptor (EGFR), epidermal growth factor receptor-3 (HER3), PD-1 , hepatocyte growth factor (HGF), insulin-like growth factor-1 receptor (IGF-1 R), and delta-like ligand 4 (DLL4).

Specifically, tivozanib is used in combination with any of chemotherapy, immunotherapy and/or radiotherapy.

Specifically, tivozanib is used in first-line therapy, or in the therapy of relapsed subjects following chemotherapy, immunotherapy and/or radiotherapy. Specifically, chemotherapy is employed with a chemotherapeutic agent, such as cisplatin or carboplatin.

Specifically, tivozanib is used adjuvant or neoadjuvant chemotherapy, optionally with further treatment, preferably with surgical resection of the tumor and/or stereotactic body radiotherapy (SBRT).

According to a specific aspect, the ROS1 positivity or the level of ROS1 expression of the cancer is determined in a biological sample containing tumor cells obtained from the subject, preferably by fluorescence in situ hybridization (FISH), immunohistochemistry (IHC), and quantitative real-time reverse transcription-PCR (qRT-PCR).

Specifically, the ROS1 positivity or expression of a tumor cell is determined by fluorescence in situ hybridization (FISH), immunohistochemistry (IHC), and quantitative real-time reverse transcription-PCR (qRT-PCR).

Specifically, the biological sample comprises any of serum, blood, faeces, tissue, a cell, urine or saliva of said human.

According to a specific aspect, ROS1 positivity is determined as ROS1 -fusion protein (e.g. by IHC) and/or ROS -fusion gene (e.g. by FISH), and/or ROS1 mRNA (e.g. by qRT-PCR), preferably indicative of a ROS1 -fusion protein selected from the group consisting of SLC34A2-ROS1 (SLC34A2 exons 13del2046 and 4 fused to ROS1 exons 32 and 34, SEQ ID 7 and 8), CD74-ROS1 (CD74 exon 6 fused to ROS1 exons 32 and 34, SEQ ID 9 and 10), EZR-ROS1 (EZR exon 10 fused to ROS1 exon 34 SEQ ID 1 1 ), TPM3-ROS1 (TPM3 exon 8 fused to ROS1 exon 35 SEQ ID 12), LRIG3-ROS1 (LRIG3 exon 16 fused to ROS1 exon 35 SEQ ID 13), SDC4-ROS1 (SDC4 exon 2 and 4 fused to ROS1 exon 32 and SDC4 exon 4 fused to ROS1 exon 34 SEQ ID 14, 15 and 16), and GOPC-ROS1 , also known as FIG-ROS1 , (GOPC exon 8 fused to ROS1 exon 35 and GOPC exon 4 fused to ROS1 exon 36) [29].

Specifically, ROS1 positivity is determined by the level and/or activity of ROSI wt and/or ROS1 ^G2932R mutant. Specifically, the subject of the present invention refers to treatment of ROS1 rearranged NSCLC.

According to a specific embodiment, the subject suffers from crizotinib or ceritinib resistance. Specifically, the subject is a patient with crizotinib-related ROS1 fusion positive tumor that acquired a ROS1 kinase domain mutation (ROS1 ^G2032R ) that confers crizotinib resistance.

Specifically, the subject is a human being, more specifically, a human patient.

Specifically, the subject suffers from a malignant epithelial tumor or adenocarcinoma, specifically wherein the tumor cells are ALK/ROS1 positive.

Specifically, the cancer is any of NSCLC, gastric cancer, serous ovarian carcinoma or cholangiocarcinoma.

Specifically, the subject suffers from early or late stage cancer. According to a specific aspect, the patient suffers advanced cancer harboring ROS1 rearrangements that is metastatic or unresectable. The invention further provides for a method for antiproliferative treatment of ROS1 positive cancer, wherein a biological sample of a patient suffering from epithelial cancer and containing tumor cells is analysed whether said tumor cells express ROS1 ; and, if the tumor cells express ROS1 , treating said patient with an effective amount of a ROS1 inhibitor, wherein said ROS1 inhibitor is tivozanib.

According to a certain aspect, the invention further provides for a method of treating or managing ROS1 positive cancer in a subject, the method comprising identifying a subject who is diagnosed with epithelial cancer and expresses ROS1 in tumor cells; and administering to the subject a therapeutically effective amount of tivozanib.

According to a further aspect, the invention provides for a method comprising: a) detecting a level of ROS1 expression and/or activity in a tumor cell obtained from a subject diagnosed with epithelial cancer; and

b) comparing the detected level to a control, wherein a detected level that is higher relative to the control identifies the cancer for treatment with tivozanib.

According to a further aspect, the invention provides for a method of treating or preventing tumor proliferation and/or metastasis, comprising:

a) detecting a level of ROS1 expression and/or activity in a tumor cell obtained from a subject diagnosed with epithelial cancer;

b) comparing the detected level to a control; and

c) if the detected level is higher relative to the control, treating the cancer with tivozanib.

Specifically, the control or control level is obtained from a reference sample, such as a sample obtained from a tumor or a cancer patient diagnosed with a tumor that is ROS1 negative, i.e. a negative control.

Specifically, the higher or elevated level as compared to a control or reference value, may be higher than a threshold or cut-off value, or higher than a reference value derived from a comparable sample. Expression may as well be determined by comparison to standards, including internal or external standards. Specifically, the term shall refer to at least a two-fold higher amount of the standard deviation, preferably at least a three-fold difference. With respect to a specific reference value, such as derived from a standard, training data or threshold, a significant increased amount is understood to refer to an at least 1 .5 fold higher amount, preferably at least 2 or 3 fold difference. According to a further spect, the invention provides for a kit for use in the treatment of ROS1 positive cancer, which comprises:

a) tivozanib; and

b) a pharmaceutically acceptable carrier,

and optionally further comprising

c) another inhibitor targeting any of ROS1 , c-MET, VEGF or RET.

Specifically, the kit is provided for the use of tivozanib according to the invention and as further described herein.

Specifically, the kit is provided for use in the treatment of a subject suffering from ROS1 positive cancer, in particular, where the subject suffers from crizotinib resistance. Specifically, the subject is a patient with crizotinib-reated ROS1 fusion positive tumor that acquired a ROS1 kinase domain mutation (ROS1 ^G2032R ) that confers crizotinib resistance. FIGURES

Figure 1. All-optical screen against RTKs and the MAPK/ERK pathway.

(a) HEK293 cells were engineered to contain an Opto-RTK (Opto-mFGFR1 , Opto-hEGFR or Opto-hROS1 ) and a MAPK/ERK pathway-responsive GFP reporter (SRE-GFP). In all-optical screening, effects of small molecules (e.g. receptor inhibitors or pathway inhibitors) are tested in 384 -well plates by first activating the RTK with blue light (λ-470 nm, I— 200 pW/cm ² ) followed by detection of pathway activity using the GFP reporter. For instance, for cells treated with inhibitors of RTKs or of components of the MAPK/ERK pathway, GFP expression will be absent. Except for small molecule addition, the process does not require contact to the cells, solution exchange or added reagents (Fig. 2). (b) Control experiments demonstrating activation of MAPK/ERK pathway by Opto-mFGFR1 and blue light as measured using the GFP reporter. Mean raw fluorescent units (RFU) ± SD (one representative experiment performed in triplicates) are shown, (c) All-optical screen against mFGFRl and MAPK/ERK pathway (68 small molecules listed in Table 1 , final concentration 5 nM). PD-166866 (final concentration 5 μΜ), a specific FGFR1 inhibitor, and DMSO were used as controls (striped and white bars). Mean percent of control (POC) values ± SEM (two independent experiments performed in triplicates) are shown, (d) Comparison of all- optical experiments with Opto-mFGFR1 (black bars) and Opto-hEGFR (white bars) allow identifying small molecules that specifically inhibit mFGFRI (ponatinib, PD- 73074), hEGFR (CI-1033, AV-412, gefitinib) or downstream proteins of the MAPK/ERK pathway (GSK-1 120212). Mean POC values ± SEM (two independent experiments performed in duplicates) are shown.

Figure 2. Workflow of the all-optical screen method (compare to Fig. 1).

Transfection can be omitted when a stable cell line is used. ^** Drug addition can be performed at the beginning of the workflow by preparing drugs in the 384-well plates. Individual steps and media are described in EXAMPLE 1 , Materials and Methods section.

Figure 3. Bioinformatics procedure for identification of orphan RTKs.

Individual steps are described in the EXAMPLE 1 , Materials and Methods section. Databased search was conducted June 2014.

Figure 4. Light activation of an orphan RTK. (a) The orphan RTK hROS1 was re-engineered to be activated by light by fusing its cytosolic domain (LBD) to the dimerizing LOV domain (LOV). Except for LCD of hROS1 , myristoylation (MYR), transmembrane (TMD), kinase (KD) and C-terminal (CTD) domains are drawn to scale (length of amino acid sequences), (b) Control experiments demonstrating activation of MAPK/ERK pathway by Opto-hROS1 and specific inhibition by crizotinib (C; final concentration 100 nM). Mean raw fluorescence units (RFU) ± SD (one representative experiment performed in triplicates) are shown.

Figure 5. All-optical screen against hROS1 and MAPK/ERK pathway. The screen contained the same 68 small molecules as in Fig. 1 (final concentration 100 nM). GSK-1 120212 (final concentration 1 μΜ) and DMSO were used as controls (striped and white bars, resp.). Mean percent of control (POC) values ± SD (two independent experiments) are shown.

Figure 6. Inhibition of the hROS1 -MAPK/ERK-axis (squares), mFGFRI - MAPK/ERK-axis (triangles) and hEGFR-MAPK/ERK-axis (spheres) by increasing doses of tivozanib. For detailed description see EXAMPLE 1 , Materials and Methods section. Mean normalized (to data point with the highest intensity in each experiment) fluorescence units (FU) ± SD (one representative experiment) are shown.

Figure 7. Inhibition of Opto-hROS1 ^G2032R by tivozanib. Control experiments demonstrating activation of MAPK/ERK pathway by Opto-h ROS 1 ^G2032R and specific inhibition by Tivozanib (T; final concentration 100 nM). Mean raw fluorescence units (RFU) ± SD (one representative experiment performed in triplicates) are shown. Fiqure 8. Inhibition of the hROS1 ^G2032R -MAPK/ERK-axis (rombs) by increasing doses of tivozanib. Experiments were performed as described in EXAMPLE 2, Materials and Methods section. Mean normalized (to datapoint with the highest intensity in each experiment) fluorescence units (FU) ± SD (one representative experiment) are shown.

Figure 9. Inhibition of hROS1 -MAPK/ERK-axis and hROS1 ^G2032R - MAPK/ERK-axis by tivozanib, crizotinib and PF-06463922. Drug final concentrations are 5 nM (black bars) and 100 nM (striped bars). DMSO (white bars) is shown as a control. Experiments were performed as described in EXAMPLE 2, Materials and Methods section). Mean POC values ± SEM from two independent experiments performed in triplicates are shown.

Figure 10: Structural modeling analysis of tivozanib and crizotinib bound to hROS1 and hROS1 ^G2032R . Comparison of the 3D structures of crizotinib (PDB ID: 3ZBF) and tivozanib (PDB ID: 4ASE) (white spheres) bound to hROS1 (PDB ID: 3ZBF) and modeled hROS1 ^G2032R . Calculated protein exposed surface is represented in black mesh. hROS1 Gly 2032 (left panels) and hROS1 modeled mutation Arg 2032 (right panels) are represented as black sticks.

Figure 11. Sequences of human ROSI wt (SEQ ID 1 ), human ROS1 ^G2932R mutant (SEQ ID 2, mutated residue marked in bold); intracellular domains used in the construction of the light-activated receptors and sequences of ROS1 fusion proteins described in uniprot and genebank databases:

SEQ ID 7: SLC34A2-ROS1 long transcript (uniprot id: M1 V485)

SEQ ID 8: SLC34A2-ROS1 short transcript (uniprot id: A9YLN5)

SEQ ID 9: CD74-ROS1 CD74 exon 6 and ROS1 exon 32 (uniprot id: M1 VPF6)

SEQ ID 10: CD74-ROS1 CD74 exon 6 and ROS1 exon 34 (uniprot id: A9YLN4) SEQ ID 1 1 : EZR-ROS1 EZR exon 10 and ROS1 exon 34 (uniprot id: J7M2B1 )

SEQ ID 12: TMP3-ROS1 TMP3 exon 8 and ROS1 exon 35 (uniprot id: M1 VPF4) SEQ ID 13: LRIG3-ROS1 LRIG3 exon 16 and ROS1 exon 35 (GenBank:

BAM95201 .1 )

SEQ ID 14: SDC4-ROS1 SDC4 exon 4 and ROS1 exon 34 (GenBank: BAM95195.1 ) SEQ ID 15: SDC4-ROS1 SDC4 exon 4 and ROS1 exon 32 (GenBank: BAM95194.1 ) SEQ ID 16: SDC4-ROS1 SDC4 exon 2 and ROS1 exon 32 (GenBank: BAM95193.1 ) DETAILED DESCRIPTION

The term "combination" with respect to treatment methods or therapy means coadministering therapeutic agents over a defined time period (e.g., weeks or months), e.g., in a concurrent manner, such as the therapeutic agents can be administered at the same or a different time and can be administered by the same route or by different routes. Combination therapies may employ administration of a therapeutic agent concurrent with surgical interventions and/or radiotherapy, e.g., the medical use o tivozanib as described herein may be combined with standard therapies, e.g. chemotherapy, immunotherapy and/or radiotherapy.

Tivozanib in combination with another therapeutic agent, e.g. a chemotherapeutic agent, may be provided in the same pharmaceutical preparation or in different pharmaceutical preparations, e.g. at the same time or at different times.

In particular, tivozanib may be provided in combination with said another therapeutic agent in a combination kit. Therefore, the invention further provides for a kit comprising one or more components to be used in combination as described herein in different containers. The kit may include, in addition tivozanib, one or more chemotherapeutic drugs, and optionally various other therapeutic agents and auxiliary agents and devices to prepare pharmaceutical formulations ready for use. A kit may also include instructions for use in a therapeutic method. Such instructions can be, for example, provided on a device included in the kit. In another specific embodiment, the kit includes tivozanib in combination with pharmaceutically acceptable carrier(s) that can be mixed before use to produce an injectable solution for near term administration.

The term "NSCLC" or "NSCL" in connection with cancer as used herein, shall refer to the term "non-small cell lung cancer" which is any type of epithelial lung cancer other than small cell lung cancer (SCLC). The term shall particularly include large cell carcinomas (LCC), adenocarcinomas (ADC), and squamous cell carcinomas (SCC).

NSCLC can be any of early stage to late stage cancer, e.g. metastatic cancer. In particular, the medical use as described herein shall refer to treatment of stages Stage 0 (carcinoma in situ), Stage I, Stage II, Stage IIIA, Stage 1MB, or Stage IV.

The term "gastric cancer" as used herein synonymously with the term "stomach cancer" shall refer to a disease in which malignant tumor cells form in the lining of the stomach. The majority of gastric cancers are adenocarcinomas. Gastric cancer is often diagnosed at an advanced stage because there are no early signs or symptoms. Surgery to remove the stomach (total gastrectomy) or part (partial gastrectomy) of the stomach is the only treatment known to cure this cancer. Radiation therapy and chemotherapy may be used after the surgery to improve the chance of a cure. In patients who cannot have surgery, radiation and chemotherapy may be used to improve symptoms and prolong survival.

The term "serous ovarian carcinoma" as used herein shall refer to a malignant neoplasm that originates from the ovary (ovarian cancer) or fallopian tube (fallopian tube cancer), and in most cases is serous epithelial cancer started in the epithelial surface layer covering the ovary or fallopian tubes. Serous ovarian carcinomas are graded into low-grade and high-grade. Low-grade serous carcinomas exhibit low-grade nuclei with infrequent mitotic figures. They evolve from adenofibromas or borderline tumors, have frequent mutations of the KRAS, BRAF, or ERBB2 genes, and lack TP53 mutations (Type I pathway). The progression to invasive carcinoma is a slow step-wise process. Low-grade tumors are indolent and have better outcome than high-grade tumors. In contrast, high-grade serous carcinomas have high-grade nuclei and numerous mitotic figures.

The term "cholangiocarcinoma" as used herein shall refer malignant neoplasm composed of mutated epithelial cells that originate in the bile ducts which drain bile from the Iiver into the small intestine. Most patients have advanced stage disease at presentation and are inoperable at the time of diagnosis. Patients with cholangiocarcinoma are generally managed with chemotherapy, radiation therapy, and other palliative care measures. These are also used as adjuvant therapies (i.e., post- surgically) in cases where resection has apparently been successful.

In ROS1 positive cancer surgery is the most common treatment for resectable tumors, is sometimes followed by chemotherapy. Chemotherapy may be given at the same time as radiotherapy, e.g., as concomitant chemoradiation or concurrent chemoradiation. Typically, radiotherapy is used after chemotherapy (sequential treatment), after surgery (adjuvant radiotherapy), or to improve cancer symptoms. Any standard cancer treatment is suitably combined with the tivozanib treatment as decribed herein.

The term "ROS1 " as used herein shall refer to the receptor tyrosine kinase which is an orphan receptor tyrosine kinase where its ligand is unknown. In the absence of a known natural ligand, ROS1 expression and/or activity of ROS1 is usually determined as follows i) fluorescence in situ hybridization (FISH), which employs a break-aparts probe specific to the ROS1 locus (e.g. ZytoLight SPEC ROS1 Dual Color Break Apart Probe); ii) immunohistochemistry (IHC), which uses ROS1 (D4D6) monoclonal antibody to detect expression levels of ROS1 protein; and iii) quantitative real-time reverse transcription-PCR (qRT-PCR), in which the total RNA is extracted form biological samples and ROS1 fusions, such as SLC34A2-ROS1 , SDC4- ROS1 , CD74-ROS1 , EZR-ROS1 , TPM3-ROS1 , LRIG3-ROS1 , GOPC-ROS1 are readily detected by PGR using a ROS1 fusion gene detection kit (e.g Amoy Diagnostics Co).

The term "ROS1 " or human ROS1 (hROS1 ) specifically shall include the ROS1 wild-type (wt) (e.g., human intracellular domain identified by SEQ ID 1 ) and/or ROS1 comprising one or more point mutations, in particular the ROS1 comprising the substitution G2032R, i.e. ROS1 ^G2032R (e.g., human intracellular domain identified by SEQ ID 2), or other mutations like the gatekeeper mutation L2026 . Some ROS1 mutations were found to confer resistance to ALK inhibitors, such as crizotanib. It has turned out that tivozanib was effectively targeting both ROSI wt and ROS1 ^G2032R

The term "ROS1 " specifically shall include the RTK or any oncogenic product thereof including e.g. ROS1 fusion genes and respective ROS1 fusion proteins. ROS1 positivity shall thus specifically include those cases where the level and/or the activity of ROS1 is determined as the level and/or activity of a ROS1 fusion gene or ROS1 fusion protein. Examples of ROS1 fusion genes and their expression products that may be determined as a measure of ROS1 fusion genes are e.g., SLC34A2-ROS1 , SDC4-ROS1 , CD74-ROS1 , EZR-ROS1 , TPM3-ROS1 , LRIG3-ROS1 and GOPC- ROS1 . Examples of ROS1 fusion proteins and their activities that may be determined as a measure of ROS1 fusion proteins are e.g. SLC34A2-ROS1 , SDC4-ROS1 , CD74- ROS1 , EZR-ROS1 , TPM3-ROS1 , LRIG3-ROS1 and GOPC-ROS1 .

The term "inhibitor" as used herein, shall refer to any compound capable of interacting with or binding to a target binding partner under conditions such that the binding partner becomes unresponsive to its natural ligands and/or is rendered inactive upon binding, thereby antagonizing the target. Inhibitors may include, small (organic) molecules, polypeptides (including e.g. proteins), peptides, nucleic acids, or oligonucleotides, in particular chemotherapeutic agents or therapeutic antibodies. The "ROS1 inhibitor" particularly antagonizes with ROS1 constitutive activation and interferes with cell pathways that cause the cancer cells to grow, form new blood vessels, and spread to other organs of the body. The goal of using the ROS1 inhibitor is to shrink the tumour and to prevent it from growing.

The term "subject" as used herein shall refer to a warm-blooded mammalian, including e.g., a human being or a non-human animal, including e.g., dogs, cats, horses, monkeys, rodents (mice and rats), and in particular includes subjects employed in invertebrates animal models (Drosophila melanogaster or Caenorhabditis elegans). In particular, the medical use of the invention or the respective method of treatment applies to a subject in need of treatment of a disease which is associated with malignant epithelial tumors identified as being ROS1 positive, or ROS1 positive cancer. The subject may be a patient suffering from disease, including early stage or late stage disease.

The term "treating" or "treatment" of a disease refers to arresting, ameliorating or inhibiting a disease (e.g. the disease, or a related disorder, condition, or symptom), reducing the risk of acquiring a disease, or reducing the development of a disease. With respect to cancer, such as malignant epithelial tumor diseases, specific treatments are employed in attempts to cure or palliate cancer. The term shall also apply to "management" of cancer, which refers to the beneficial effects that a patient derives from a therapy, which does not result in a cure of cancer, but may prevent the progression or worsening of the cancer.

The term "therapeutically effective amount" or "effective amount" as used herein shall refer to the amount of a compound that, when administered to a subject for treating a disease, or at least one of the clinical symptoms of a disease, provides a therapeutic benefit or is sufficient to affect such treatment of the disease or symptom thereof, e.g., sufficient to delay or minimize the spread of cancer, in particular late stage cancer. The term "effective amount" shall particularly refer to amelioration of symptoms associated with malignant epithelial tumors, also encompassing an amount that improves overall therapy, reduces or avoids unwanted effects, or enhances the therapeutic efficacy of or synergizes with one or more antineoplastic agents, utilized in combination therapies as described herein.

The therapeutically effective amount may vary depending, e.g., on the compound, the disease and/or symptoms of the disease, severity of the disease and/or symptoms of the disease or disorder, the condition of the patient and the route of delivery. An appropriate amount may be readily ascertained in accordance with routine pharmacological procedures. The term "tivozanib" as used herein shall refer to the compound also known as AV-951 and KRN951 , with the following formula: 1 -{2-Chloro-4-[(6,7- dimethoxyquinolin-4-yl)oxy]phenyl}-3-(5-methylisoxazol-3-yl) urea and having the chemical structure:

The term shall particularly include pharmaceutically acceptable salts, solvates, solvates of a pharmaceutically acceptable salt, esters, or polymorphs thereof. In certain embodiments, tivozanib is or 1 -{2-Chloro-4-[(6,7-dimethoxyquinolin-4- yl)oxy]phenyl}-3-(5-methylisoxazol-3-yl)urea or hydrates of a hydrochloride salt. In certain embodiments, tivozanib is or 1 -{2-Chloro-4-[(6,7-dimethoxyquinolin-4- yl)oxy]phenyl}-3-(5-methylisoxazol-3-yl)urea monohydrochloric acid salt monohydrate.

"Pharmaceutically acceptable" refers to approved or approvable by a regulatory agency of the European Union or U.S. Federal or a state government or listed in the generally recognized pharmacopoeia for use in animals, and more particularly in humans.

A pharmaceutically acceptable salt is herein understood as a parent compound, which possesses the desired pharmacological activity of the parent compound. Such salts include acid addition salts, formed with inorganic acids and one or more protonable functional groups such as primary, secondary, or tertiary amines within the parent compound. In certain embodiments the salts are formed with organic acids such as acetic acid, propionic acid, hexanoic acid, cyclopentanepropionic acid, glycol ic acid, pyruvic acid, lactic acid, malonic acid, succinic acid, malic acid, maleic acid, fu marie acid, tartaric acid, citric acid, benzoic acid, 3-(4-hydroxybenzoyl) benzoic acid, cinnamic acid, mandelic acid, methanesulfonic acid, ethanesulfonic acid, 1 ,2-ethane- disulfonic acid. Pharmaceutically acceptable salts may be hydrates or other solvates, as well as salts in crystalline or non-crystalline form.

Tivozanib is particularly provided in a pharmaceutical preparation, such as in an oral preparation, e.g., including tablets, capsules, powders, granules, and syrups, or a parental preparation, e.g., including injections (infusion preparations), suppositories, tapes, and ointments. These various preparations may be prepared by conventional methods, for example, with commonly used pharmaceutically acceptable carriers, such as excipients, disintegrants, binders, lubricants, colorants, and diluents. Excipients include, for example, lactose, glucose, corn starch, sorbit, and crystalline cellulose; disintegrants include, for example, starch, sodium alginate, gelatin powder, calcium carbonate, calcium citrate, and dextrin; binders include, for example, dimethylcellulose, polyvinyl alcohol, polyvinyl ether, methylcellulose, ethylcellulose, gum arabic, gelatin, hydroxypropylcellulose, and polyvinyl pyrrolidone; lubricants include, for example, talc, magnesium stearate, polyethylene glycol, and hydrogenated vegetable oils. Liquid formulations can be solutions, emulsions or suspensions and can include excipients such as suspending agents, solubilizers, surfactants, preservatives, and chelating agents. Injections or infusion preparations contain, for example, pharmaceutically acceptable carriers including e.g, sterile water, saline, phosphate buffered saline, dextrose, glycerol, ethanol, and the like, as well as combinations of any thereof.

Pharmaceutically acceptable carriers suitable used to formulate the pharmaceutical preparation are known in the art and described in, e.g. REMINGTON'S PHARMACEUTICAL SCIENCES.

The content of tivozanib in the pharmaceutical composition may vary depending on the dosage form. In general, however, the content is 0.5 to 50% (w/w), preferably 1 to 20% (w/w), based on the whole composition.

Preferably, tivozanib is administered as an oral tablet or capsule or as an intravenous (iv) bolus injection or infusion.

When administered as an oral tablet or capsule, the dosage of tivozanib may be a single capsule or tablet or two or more capsules or tablets.

The preparation is typically administered in an amount of 0.01 to 100 mg/kg, preferably 0.1 to 50 mg/kg. This dose can be administered once a day or divided doses of several times daily. A preferred dose is e.g. within the range of 0.5-2.0 mg daily, preferably at 1 .5 mg daily.

Tivozanib may be administered on a repeating schedule of one dose per day for one or more weeks, e.g. three weeks, followed by a period where tivozanib is not administered, e.g. one week off.

In routine screening experiments, tivozanib was previously not found to effectively inhibit ALK, thus, it was not considered as a potential second-generation ALK inhibitor. Therefore, tivozanib has unexpectedly been identified as a potent inhibitor of ROS1 and in particular ROS1 ^G2032R . This enables the molecularly targeting therapy against ROS1 positive cancer, without the side effect profile typically expected when using an ALK inhibitor.

Following an optogenetics-assisted , cellular-based screening approach that allows screening for inhibitors against several RTKs (FGFR and EGFR) and the mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK), tivozanib was found to be a selective inhibitor of the orphan ROS1 . Engineered human cells containing engineered blue light-activated RTKs (Opto-RTKs, [30]) and a genetic fluorescent MAPK/ERK pathway reporter (SRE-GFP) was used (see figure 1 A). This screening platform uses light for activation and detection of cell signaling, therefore obviating the need for the addition of reagents, limiting the number of operational steps and providing new experimental strategies to increase specificity and counter variability. Because activating ligands are not required, orphan RTKs, such as ROS1 , could be a target for optogenetics-assisted drug screening given that their catalytic domains can be separated from the ligand-sensing domains [31 , 32].

Therefore, a blue light-activated hROS1 (opto-hROS1 ) was prepared, applying the same approach developed previously [30]. Notably, the optical control of an orphan receptor could be shown for the first time. Using this optical drug screening platform, 68 small molecules were screened in the 384-well plate format and three molecules that gave a "hit" in an experiment with Opto-ROS1 were found. Two of these molecules were known kinase inhibitors of ROS1 or components of the MAPK/ERK pathway: crizotinib and GSK-1 120212, respectively. Surprisingly, a new molecule that inhibited light-activated ROS1 : tivozanib (AV-951 ), was identified (Figs. 5 and 6. SC ₅₀ : ~10 nM). Next, the mutation G2032R was engineered in the kinase domain of the opto-hROS1 and tivozanib was also found to robustly inhibit this mutant (IC50: ~50 nM) (Figs. 7 and 8). When then tivozanib was compared to other known inhibitors of hROSI wt and/or hROS1 ^G2032R (e.g. crizotinib and PF-06463922) at two different drug concentrations, tivozanib was found to inhibit hROS1 ^G2032R more efficiently than crizotinib and PF- 06463922 (Fig. 9).

Therefore, tivozanib is provided as selective ROS1 inhibitor for use as molecularly targeted therapy. In particular, drug resistance due to the emergence of a ROS1 mutant (eg. ROS1 ^G2032R ) may be effectively treated, because tivozanib was found to inhibit both ROS1 and ROS1 ^G2032R activity using the optogenetics-assisted, cell-based drug screening method [33-35]. As tivozanib is safe (it is currently undergoing clinical trial investigation for other indications), it may be advanced as a new therapeutic candidate for treating patients with ROS1 -driven malignancies, such as non-small cell lung cancer, gastric cancer, serous ovarian carcinoma or cholangiocarcinoma.

The foregoing description will be more fully understood with reference to the following example. Such examples are, however, merely representative of methods of practicing one or more embodiments of the present invention and should not be read as limiting the scope of invention.

EXAMPLES

EXAMPLE 1 : Light activation of cell signaling permits all-optical small molecule screening

High-throughput live-cell screens are intricate elements of systems biology studies and drug discovery pipelines. Here, an optogenetics-assisted method is demonstrated that obviates the addition of chemical activators and reporters, reduces and the number of operational steps in a cell-based small molecule screen against protein kinases including the Orphan' ROS1 receptor tyrosine kinase. This blueprint for all-optical screening can be adapted also to other drug targets and cellular processes.

Over the past decades, many chemical processes have been improved by replacing additives, such as catalysts, initiators or emulsifiers, with physical stimuli, such as light or ultrasound [36-39]. Although replacement resulted in reduced cost, increased robustness and improved sustainability, this general principle has not yet found many adaptations in chemical biology. Automated screens using living cells are essential in the identification and characterization of small molecules that act on disease-related proteins and cellular pathways. However, in many cell-based screens the need to add reagents that alter or report on cell activity results in complex operational design, high cost and sources of error. Furthermore, mammalian cells are sensitive to environmental perturbations (e.g. temperature or ionic strength) and subject to inherent variability. In neurobiology and cell biology, optogenetics and photopharmacology have recently harnessed the power of light to manipulate the behavior of cells and animals non-invasively and with high spatial and temporal precision [40-42]. Here, an optogenetics-assisted , cell-based screening method was developed that interrogates receptor tyrosine kinases (RTKs) and the mitogen- activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) pathway comprising one G-protein (Ras) and three intracellular kinases (Raf, MEK and ERK). It was demonstrated that in this screening method the use of light for activation and detection of cell signaling obviates the need for the addition of reagents, limits the number of operational steps and provides new experimental strategies to increase specificity and counter variability.

The MAPK/ERK pathway is activated by RTKs and regulates cell survival, proliferation and differentiation. Modulators of the MAPK/ERK pathway, of RTKs and of other protein kinases are intensively pursued as new therapeutics in cancer and metabolic and neurodegenerative disorders. First human embryonic kidney 293 (HEK293) cells were engineered that contain light-activated RTKs and a genetic fluorescent MAPK/ERK pathway reporter (Fig. 1a). The light-activated RTKs (also called Opto-RTKs') are modified growth factor receptors that are insensitive to their natural ligands but activated by blue light-induced homodimerization through incorporation of the light-oxygen-voltage-sensing (LOV) domain of aureochromel from V. frigida [43]. Two Opto-RTKs, the light-activated murine fibroblast growth factor receptor 1 (Opto-mFGFR1 ) and the light-activated human epidermal growth factor receptor (Opto-hEGFR), were initially employed to set-up this platform.

In the genetic fluorescent reporter (SRE-GFP), tandem repeats of serum response element (SRE)[44], an enhancer sequence that is responsive to cell signaling pathways including the MAPK/ERK pathway, precede a gene coding for the green fluorescent protein (GFP). The engineered cells respond to light stimulation at a wavelength and intensity suited for Opto-RTK activation (λ~470 nm, intensity-200 pW/cm ² ) with increased production of GFP (Fig. 1 b), and the combination of Opto- RTKs and the GFP reporter thus enables a novel 'all-optical' mode of operation where light is used to activate as well as to read cellular signaling.

Using this mode of operation, small molecules were screened in the 384 -well plate format without the addition of reagents to induce or detect pathway activation and with a small number of handling steps (Fig. 1c and Fig. 2) (Z'-factor > 0.7, see Materials and Methods and Table 3). In addition, all wells in the 384-well plate can be activated at the same time and with nearly identical intensity using light emitting diodes (LEDs) (deviation of intensity was <0.05% over the entire plate with a day-to- day variability of <0.05%). In a set of kinase inhibitors (Table 1 ), three molecules inhibited the mFGFRI -MAPK/ERK-axis by >50%, and these molecules were known kinase inhibitors of FGFR1 or components of the MAPK/ERK pathway (Fig. 1c, d). To demonstrate the ability of the method to identify whether ligands act at the receptor or downstream pathway, the same screen was conducted with Opto-hEGFR. Unlike FGFR1 , which couples to the MAPK/ERK pathway via additional FGFR substrate (FRS) adaptor proteins, EGFR directly activates the pathway. Three molecules inhibited the EGFR1 -MAPK/ERK-axis, and these molecules were known kinase inhibitors of EGFR or components of the MAPK/ERK pathway (Fig. 1d). Because inhibitors of FGFR1 were not detected in the screen with EGFR and vice versa, the method can selectively identify those small molecules that specifically act on a receptor or the common downstream proteins. Collectively, these results demonstrate a cell-based small molecule screen in the 384-well format assisted by optogenetics.

This method functions without added reagents (e.g. peptide ligands or detection assays) and a reduced number of operational steps as the method does not require physical contact to the living cells. Furthermore, because activation by a peptide or other agonist is not required, the all-optical method has the capability to conduct screens against 'orphan' receptors, i.e. receptors for which native ligands are not known. To demonstrate the optical control of an orphan receptor, a de novo database search and bioinformatics analysis was first conducted to identify human orphan RTKs (Fig. 3 and Materials and Methods). The search reported human ROS1 (hROS1 ), a proto-oncogene orphan RTK that is activated by fusion with other proteins in a variety of tumor cell types [13, 45]. To engineer a light-activated variant of human hROS1 (Opto-hROS1 ), the dimerizing LOV domain was fused to the intracellular domain of hROS1 (Fig. 4 and Materials and Methods). When testing the kinase inhibitor library against Opto-hROS1 , three molecules that inhibited the hROS1 -MAPK/ERK-axis were found (Fig. 5). Two of these molecules (crizotinib and GSK-1 120212) were known kinase inhibitors of hROS1 and components of the MAPK/ERK pathway, resp. The third molecule (AV-951 ) was active against hROS1 but not hEGFR or mFGFRI (Fig. 6). Notably, AV-951 was previously not assigned to inhibit hROS1 [46].

In summary, it was demonstrated that incorporation of optogenetics enables cell-based small molecule screens without the use of additives and with a minimum number of operational steps. This work is the first expansion of optogenetics into the field of drug discovery and demonstrates potential for high-throughput systems biology as well. Here, light acted as a universal ligand for receptors of different families (e.g. FGFR1 , EGFR or ROS1 ), yet, at the same time, activation was highly specific as only the genetically-engineered receptors were stimulated. Interference from endogenous receptors that may be also activated by added ligands can be excluded, which is particularly desirable when targeting receptors for which ligands with sufficient specificity are not available. For instance, in the experiments with RTKs performed above, EGF or FGF2 will have bound to several receptor proteins that are expressed in this cell type (FGFR1 , FGFR2, FGFR3, FGFR-like 1 and EGFR), while Opto-RTK activation was specific to the receptor that was engineered. Activation by light in real time may also reveal more detailed insights into molecular inhibition and cellular signaling mechanisms. For instance, the interaction between small molecules and proteins may be activation-state dependent [47, 48], and the duration and frequency of activation may determine the choice of pathway or functional outcome [49, 50]. The ability to switch signals on with temporal precision and tunable strength, even within the same well, may be used to explore these phenomena in a systematic and automated manner. This could be of particular use when exploring 'mutagenesis space' in the form of a library of modified receptors. Indeed, the method is well suited for experiments with genetic libraries as is compatible with transient cell transfection. This approach can be extended and adapted to other drug targets because of the increasing number of available optogenetic tools and fluorescent reporters.

Materials and Methods

Kinase inhibitor library

Small molecules are listed in Table 1 . Out of the tested 68 small molecules, 62 molecules target protein kinases.

Gene constructs

Opto-mFGFR1 and Opto-hEGFR in pcDNA3.1 (-) (Life Technologies) were described previously [43]. Identification of hROS1 and genetic engineering of Opto- hROS1 is described below. The SRE-GFP reporter vector was obtained from Qiagen/SA Biosciences.

Cell culture and transfection

HEK293 were derived by F.L. Graham (McMaster University). HEK293 cells were maintained in DMEM, resp., in a humidified incubator with 5% CO2 atmosphere. Media was supplemented with 10% FBS, 100 U/ml penicillin and 0.1 mg/ml streptomycin. For transfection of HEK293 cells, 2x10 ⁶ cells were seeded in 60 mm cell culture dishes coated with poly-L-ornithine (PLO, Sigma). Cells were transfected with 4.04 to 8.04 pg total DNA per dish (receptor, pcDNA3.1 (-) empty vector, and reporter at a ratio of 1 :50:50 or 1 :100:100) using polyethylenimine (Polysciences). For mock transfected cells (Fig. 1 b ΉΕΚ293"), receptor vector was omitted in the transfection mixture.

Custom incubator for light stimulation of well plates

A thermoelectric incubator (PT2499, ExoTerra) was equipped with 300 RGB LEDs (5050SMD, X _ma x * 630 nm (red light), X _max * 530 nm (green light), X _max ~ 470 nm (blue light), bandwidth ¾ ±5 nm). Light intensity was controlled with a dimmer and measured with a digital power meter (PM120VA, Thorlabs). Blue light intensity at maximal output was 247 iiW/cm ² . Light of this intensity is sufficient for activation and well tolerated by mammalian cells without signs of toxicity even for extended periods of time. Hardware to maintain a CO ₂ atmosphere is not required if medium supplemented with HEPES (25 mM) is used during light stimulation (see below). To measure profile of light distribution, the sensor of the power meter was mounted on a holder and moved in 1 cm steps. Intensity was recorded and deviation between highest and lowest intensity was calculated (Δ<0.13 iiW/cm ² ). Day-to-day variability of light was measured in the same way but on several days distributed over one week (Δ<0.13

All-optical drug screening against RTKs and the MAPK/ERK pathway

The workflow is depicted in Fig. 2. Transfected HEK293 cells were kept in DMEM (supplemented with 5% FBS and no antibiotics; "D5-AB" medium) for 6 h. Afterwards, 5Ό00 to 20Ό00 cells were seeded in each well of 384-well plates (3712, Corning) in low-fluorescence medium (FlouroBrite ^™ , Life Technologies, supplemented with 25 mM HEPES, 0.5% FBS, 100 U/ml penicillin and 0.1 mg/ml streptomycin, pH 7.5; "COI" medium). Small molecules were added and after 1 h cells were stimulated with blue light in a custom incubator (see above). Unstimulated cells were kept in the dark by covering selected wells on the same 384-well plate. GFP fluorescence was then measured in a microplate reader (Synergy H1 , BioTek) at the optimized excitation wavelength of 500±5 nm and emission wavelength of 535±5 nm (10 measurements per well, measurement duration: 10 ms, gain: 90 to 130). Identification of Orphan RTKs

Orphan RTKs were identified using the bioinformatics procedure described in Fig. 3. Protein family search with the PFAM motif "Pkinase_Tyr" (PF07714) at the Wellcome Trust Sanger Institute (http://pfam.xfam.org/)[51 ] retrieved a comprehensive list of human Tyr kinases. This PFAM motif is a good representative of kinase domains found in RTKs (it was confirmed that it is found in members of diverse RTK families, such as fibroblast growth factor receptors, ErbB receptors, Insulin-like growth factor receptor, neurotrophin receptors, ROR receptors). Retrieval of these sequences was followed by transmembrane helix prediction with TMHMM [52] to retain only sequences with transmembrane helices. To remove sequence redundancy and assign sequence fragments clustering with UCLUST [53] was performed. 69 clusters with a unique candidate sequence each for RTKs were obtained Table 2 and orphan RTKs were identified by manual curation.

Genetic engineering of Opto-ROS1

A sequence coding for the ROS1 gene was obtained from the Mammalian Gene

Collection (Dharmacon, GE Life Science). Using inverse PGR, an expression vector was prepared starting from an Opto-mFGFR1 vector in which the mFGFRI ICD was replaced by two inverted Sapl restriction sites. The ROS1 ICD was amplified with oligonucleotides (F: GAT CGC TCT TCA GAG CAT AGA AGA TTA AAG AAT CAA AAA AG (SEQ ID 3), R: GAT CGC TCT TCC AGG ATC AGA CCC ATC TCC ATA TCC ACT G (SEQ ID 4)) and PGR and inserted into the vector using 'Golden Gate' cloning [54].

Assay validation and statistical analysis

Z' factors (Table 3) were calculated for interleaved-signal format plates following

3 x (std (c ₊ ) + std (c_))

Z factor = 1 ^■ — ₍

\ vg(c ₊ ) - avg(c_) \

where std (c+) and avg {c ₊ ) are the standard deviation and the average of

DMSO treated samples, resp., and std (c_) and avg{c_) are the standard deviation and the average of PD-166866 (INH; final concentration 20μΜ) treated samples, resp. On each plate, ten columns of DMSO treated samples and ten columns of IHN treated samples were tested in groups of two and in alternating order). Z factors for these plates were >0.70 (Table 3) and failures were not observed. Percentage of control (POC) values were calculated following: Xt

POC = —- x 100

avg{c+)

where X _t is the measurement of the i ^th small molecule and avg(c ₊ ) is the average measurement of the DMSO treated samples.

Table 1 : Small molecules used in this example. Number corresponding to compounds in Fig. 1 c and Fig. 5.

# Small molecule Canonical target(s) Source

1 R-406 Syk MedChem Express

2 GDC-0941 ΡΙ3Κα/δ Selleck

3 GSK-1 120212 MEK1/2 Selleck

4 Sunitinib VEGFR2, PDGFR , c-Kit Synthesis

5 Imatinib v-Abl, c-Kit, PDGFR Selleck

6 Erlotinib EGFR Synthesis

7 Vargatef VEGFR1 /2/3, FGFR1 /2/3, PDGFRa/β Synthesis

8 CI-1033 EGFR, ErbB2 Selleck

9 MP-470 c-Kit, PDGFRa. Flt3 Selleck

10 Flavopiridol CDK1 ,'2/4/6, EGFR, PKA Santa Cruz

1 1 Ponatinib Abi, PDGFRa, VEGFR2, FGFR1. Src ARIAD

12 AT-9283 JAK2/3 Selleck

13 Paclitaxel Microtubule polymer stabilizer Sigma

14 SR-3677 ROCK2 Sigma

15 BX-795 PDK1 Selleck

16 SB-202190 ρ38α/β Selleck

17 BX-912 PDK1 Selleck

18 Purvalanol B CDC2, CDK2/4/5 Tocris

19 AG-13958 VEGF Synkinase

20 Ki-20227 c-Fms, VEGFR2, PDGFR , c-Kit APExBio

21 Crizotinib c-Met, ALK APExBio

22 asatinib Abl, Src, c-Kit, Synthesis

23 Sorafenib Raf-1 , B-Raf, VEGFR-2 Synthesis

24 Lapatinib EGFR, ErbB2 Synthesis

25 CYT1 1387 JAK1 /2 Synthesis

26 Temsirolimus mTOR Selleck

27 Bosutinib Src, Abl Synthesis

28 CP-724714 ErbB2 Selleck

29 AMG-Tie2-1 Tie-2 Synkinase

30 TAK-715 p38a Selleck

31 LDE225 SMO Selleck

32 SB-590885 B-Raf APExBio

33 TG-101209 JAK2, F!t3, RET Selleck

34 Rho-15 ROCK1 /2 Synthesis

35 Dasatinib Abl, Src, c-Kit Synthesis

36 Gefitinib EGFR Synthesis Mosetanib VEGFR1 /2/3 Synthesis

TGX-221 ρ1 10β Selieck

Sanguinarine Na' K'-and Mg ⁺ -ATPase inhibitor Santa Cruz

LDN-57444 Proteasome inhibitor for Uch-L1 Selieck

PF-477736 Chk1 , VEGFR2, Aurora-A. FGFR3, Flt3, Fms, Ret, Yes Sigma

Lenvatinib VEGFR2/3/1 , FGFR1 , PDGFRa/β Source

FR180204 ERK1 /2 MedChem Express

KU-0063794 mTORC1/2 Selieck

Nilotinib Bcr-Abl Selieck

JNJ-7706621 CDK1 /2, Aurora A/B Synthesis

PF-62271 FAK, Pyk2 Selieck

GW-441 56 TrkA Synthesis

Oligomycin ATP synthase Synthesis

GSK-269962 ROCK1 /2 Selieck

ER-27319 Syk Selieck

Vandetanib VEGFR2 Santa Cruz

Enzastaurin ΡΚΟβ/α/γ/ε ARIAD

Rapamycin mTOR Selieck

PD-04217903 c-Met Sigma

AT-7519 CDK1 /2/4/6/9 Sigma

RWJ-67657 ρ38α/β Selieck

TG-101348 JAK2/1 /3, BRD4 Selieck

GSK-690693 Akt1/2/3, PKA, PrkX, PKC Selieck

Bay 1 1 -7082 NF-KB, ΙκΒα Tocris

PD-173074 FGFR1 Synkinase

Tandutinib FLT3, PDGFR, c-Kit APExBio

ABT-737 Bcl-xL, Bcl-2, Bcl-w APExBio

AV-951 VEGFR1 /2/3, PDGFR, c-Kit Synthesis

B S-2 c-Met Synthesis

AV-412 EGFR, ErbB2 Synthesis

ZSTK-474 PI3K6 Synthesis

Table 2: Seed sequences retrieved using the bioinformatics procedure described in Fig. 3. Where necessary for clarity, gene short names were added in square brackets to the Uniprot description. Stars (*) denote RTKs identified as orphans.

Uniprot-ID Uniprot Description

D7RF68 AGTRAP-BRAF fusion protein

Q59HE0 Colony stimulating factor 1 receptor variant

Q6P4R6 EPH receptor A3

Q7Z635 EPH receptor B4

B7ZKW7 EPHA5 protein

Q4LE53 EPHB2 variant protein

P21709 Ephrin type-A receptor 1

E7EML7 Ephrin type-A receptor 10

P29317 Ephrin type-A receptor 2

P54764 Ephrin type-A receptor 4

Q9UF33 Ephrin type-A receptor 6

D6RAL5 Ephrin type-A receptor 6

P29322 Ephrin type-A receptor 8

P54762 Ephrin type-B receptor 1

P29323 Ephrin type-B receptor 2

P54753 Ephrin type-B receptor 3

015197 Ephrin type-B receptor 6

A2ABM8 Epithelial discoidin domain-containing receptor 1

E7EU09 Fibroblast growth factor receptor [FGFR1 ]

D3DRD5 Fibroblast growth factor receptor [FGFR2]

F8W9L4 Fibroblast growth factor receptor [FGFR3]

P22607 Fibroblast growth factor receptor 3

Q59F30 Fibroblast growth factor receptor 4 variant

P25092 Heat-stable enterotoxin receptor

P08581 Hepatocyte growth factor receptor

B2RE75 Highly similar to Homo sapiens c-mer proto-oncogene tyrosine kinase (MERTK)

A8K2T7 Highly similar to Homo sapiens epidermal growth factor receptor (erythroblastic leukemia viral (v-erb-b) oncogene homolog avian) (EGFR)

A8KAM8 Highly similar to Homo sapiens platelet-derived growth factor receptor betapolypeptide

(PDGFRB)

E9PFZ5 Inactive tyrosine-protein kinase 7

P06213 Insulin receptor

P14616 Insulin receptor-related protein

P08069 Insulin-like growth factor 1 receptor

Q59EB0 Kinase insert domain receptor (A type III receptor tyrosine kinase) variant [VEGFR2]

P29376 Leukocyte tyrosine kinase receptor P07333 Macrophage colony-stimulating factor 1 receptor

Q04912 Macrophage-stimulating protein receptor

P10721 Maststem cell growth factor receptor Kit

F5H3K9 Muscle skeletal receptor tyrosine-protein kinase

Q16288 NT- 3 growth factor receptor

P16234 Platelet-derived growth factor receptor alpha

Q9H5K3 ^* Protein kinase-like protein SgK196

P08922* Proto-oncogene tyrosine-protein kinase ROS

Q6MZT2 Putative uncharacterized protein DKFZp686D1354 [DDR2]

A1 L4F5 Receptor tyrosine kinase-like orphan receptor 2 [ROR2J

P04626 Receptor tyrosine-protein kinase erbB-2

P21860 Receptor tyrosine-protein kinase erbB-3

Q15303 Receptor tyrosine-protein kinase erbB-4

Q5VTU6 Receptor-type tyrosine-protein kinase FLT3

F8TLW0 Ret proto-oncogene tyrosine-protein kinase receptor isoform a

Q02846 Retinal guanylyl cyclase 1

P51841 Retinal guanylyl cyclase 2

Q8IWU2 Serinethreonine-protein kinase LMTK2

Q96Q04 Serinethreonine-protein kinase LMTK3

Q59HG2 TEK tyrosine kinase variant

Q59FM9 TYR03 protein tyrosine kinase variant

Q15516 Tyrosine kinase [Mer variant]

F8WED1 Tyrosine-protein kinase Mer

F0UY65 Tyrosine-protein kinase receptor [Alk]

A9YLN4 Tyrosine-protein kinase receptor [CD4-ROS1 fusion]

A9YLN5 Tyrosine-protein kinase receptor [SLC34A2-ROS1 fusion]

A8K3Z4 Tyrosine-protein kinase receptor [trkA]

Q548C2 Tyrosine-protein kinase receptor [trkBJ

P35590 Tyrosine-protein kinase receptor Tie-1

F5H4Q1 Tyrosine-protein kinase receptor TYR03

P30530 Tyrosine-protein kinase receptor UFO

P34925 Tyrosine-protein kinase RYK

Q6J9G0 ^* Tyrosine-protein kinase STYK1

A2VCQ3 Tyrosine-protein kinase transmembrane receptor ROR1

P35916 Vascular endothelial growth factor receptor 3 Table 3: Statistical analysis of assay data. Statistics were determined for interleaved-format plates (each plate contained ten columns for DMSO and ten columns for IHN with alternating order; also see Materials and Methods).

Plate Statistics

Day 1 , Plate 1 Mean (DMSO) 40756

SD (DMSO) 3689

Mean (IHN) 1217

SD (IHN) 271

Z 0.70

Day 1 , Plate 2 Mean (DMSO) 42404

SD (DMSO) 3925

Mean (IHN) 1225

SD (IHN) 269

Z 0.69

Day 2, Plate 1 Mean (DMSO) 60338

SD (DMSO) 5068

Mean (IHN) 1632

SD (IHN) 341

Z 0.72

Day 2. Plate 2 Mean (DMSO) 53434

SD (DMSO) 4865

Mean (IHN) 1599

SD (IHN) 318

Z 0.70

EXAMPLE 2: Tivozanib is a potent inhibitor against a crizotinib-resistant human ROS1 mutant.

Lung cancer is the leading cause of cancer deaths worldwide. Indeed, lung cancer is a complex group of diseases that encompasses two major histological tumor types, non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). NSCLC represents 80% of all lung cancers and includes several subtypes based on the presence of selective mutations. The most frequent mutations include those present in the genes encoding K-RAS, EGFR, ALK, B-RAF, HER2, PI3K, RET and ROS1 [55]. Most of these mutant genes encode drug-target kinases, a property that has allowed the development of selective drugs such as gefitinib and erlotinib for tumors with mutations in EGFR and crizotinib for ALK- and ROS1 -driven tumors. These drugs are now been widely used to treat patients carrying these tumors subtypes [55]. Unfortunately, patients treated with these drugs occasionally develop resistant mutants of the target kinases. In particular, a subset of ROS1 -driven tumor patients treated with crizotinib acquired a ROS1 mutation (ROS1 ^G2032R ) that confers resistance against this drug [19, 20]. Therefore, there is an urgent need to develop novel and efficient medicines to treat these resistant tumors.

The study described in EXAMPLE 1 was therefore extended and the effect of tivozaninb was tested on the ROS1 ^G2032R crizotinib-resistant mutant. The mutation G2032R (number corresponding to the hROS1 wt protein) was engineered in the kinase domain of the opto-hROS1 . Tivozanib was found to strongly inhibit this mutant receptor at a final concentration of 100 nM (Fig. 7) and in a dose-dependent manner (IC50: ~50 nM, Fig. 8). Then, tivozanib was compared to other known hROS1 and/or hROS1 ^G2032R inhibitors, such as crizotinib and PF-06463922 (Table 4) at two different concentrations (5nM and 100nM). Using an optogenetics-assisted platform, tivozanib was found to block hROS1 and hROS1 ^G2032R activity more efficiently than crizotinib and PF-06463922 (Fig. 9), respectively.

Next, crizotinib and tivozanib atomic interactions with ROS1 and ROS1 ^G2032R were compared through structural modeling analysis. In hROS1 , both drugs can access easily to the drug binding pocket (Fig. 10), therefore inhibiting the receptor. However, in the ROS1 ^G2032R variant, the mutated Arginine is acting as a "gate-keeper" mutation, impeding the access of crizotinib to the binding pocket (Fig 10); while in case of tivozanib, the flat shape of the molecule can overcome this restriction and enter into the cavity. Remarkably, this result indicates that flatter side-groups are able to lay under "gate-keepers" residues, and therefore this strategy could be use to engineer new kinase inhibitors.

Furthermore, the pharmacokinetic and toxicity parameters of tivozanib and crizotinib were compared (table 5). The result of this comparison indicated that tivozanib offers i) longer half-life (~5 days), ii) a lower mean plasma concentration (-1 14 ng/ml), but enough to inhibit both ROS1 and ROS1 ^G2032 (IC50: 10 and 50 nM, resp.); iii) lower clearance (-0.60 L/h), which enables to reduce the dosage, therefore enhancing the availability and prolonging the half-life of the small molecule; iv) reduced dose in humans (1 .5 mg), which reduces side effects and increase compliance for the patient and v) lower discontinuation percentage (37%).

In summary, tivozanib is provided as a selective ROS1 and ROS1 ^G2032R inhibitor for use as molecularly targeted therapy. The structural modeling analysis provided molecular basis of the drug interaction with both targets, which explained the higher inhibition potency. Importantly, tivozanib is safe in humans and has superior pharmacokinetics and toxicity properties compared to crizotinib. Therefore, tivozanib may be advanced as a new therapeutic candidate for treating patients suffering ROS1 - driven non-small cell lung cancer and other tumors in which ROS1 plays an important role, such as gastric cancer, serous ovarian carcinoma or cholangiocarcinoma.

Materials and Methods

Genetic engineering of Opto-ROS1 ^R2032R

Glycine 2032 in Opto-hROS1 (see EXAMPLE 1 Materials and Methods) was mutated to Arginine by site directed mutagenesis using oligonucleotides (AAT AAG TAA GAA GGT CTC TTC CCT CCA TCA GTT CCA G (SEQ ID 5) and CTG GAA CTG ATG GAG GGA AGA GAC CTT CTT ACT TAT T (SEQ ID 6)). Mutation was checked by DNA sequencing.

Cell culture and transfection

HEK293 cells were maintained in DMEM, in a humidified incubator with 5% CO2 atmosphere. Media was supplemented with 10% FBS, 100 U/ml penicillin and 0.1 mg/ml streptomycin. For transfection, 2x10 ⁶ cells were seeded in 60 mm cell culture dishes coated with poly-L-ornithine (PLO, Sigma). Cells were transfected with 4.04 to 8.04 g total DNA per dish (receptor, pcDNA3.1 (-) empty vector, and reporter at a ratio of 1 :50:50 or 1 :100:100) using polyethylenimine (Polysciences).

Small molecule testing against hROS1 and hROS1 ^G2032R and the MAPK/ERK pathway

For drug dose response curves and individual drug test, cells were transfected with the indicated opto-RTKs and seeded in 384 well plates as described in EXAMPLE 1 Materials and Methods. Then, small molecules tivozanib, crizotinib and PF- 06463922 (Table 4) were added at indicated final concentrations. Samples were subsequently stimulated as described in EXAMPLE 1 Materials and Methods.

Structural modeling analysis

Tridimensional atomic structures of tivozanib bound to VEGF receptor (PDB ID:

4ASE) and crizotinib bound to hROS1 receptor (PDB ID: 3ZBF) were selected. Structures were superposed and an overall RMSD value of 2.0 and 39% identity was calculated using the DALI server (http://ekhidna.biocenter.helsinki.fi). This indicates that compared structures are highly similar and, therefore, they could be used for drug interaction modeling. Structures were then superposed and Glycine 2032 was substituted by Arginine to model hROS1 ^G2032R using Pymol (www.pymol.org).

Table 4: Small molecules used in this example. Number corresponding to compounds in Fig. 8 and Fig. 9.

Table 5: Pharmacokinetic and toxicity properties for tivozanib and crizotinib.

REFERENCES

1 . Acquaviva, J., R. Wong, and A. Charest, The multifaceted roles of the receptor tyrosine kinase ROS in development and cancer. Biochim Biophys Acta, 2009. 1795(1 ): p. 37-52.

2. Birchmeier, C, et al., Characterization of an activated human ros gene. Mol Cell Biol, 1986. 6(9): p. 3109-16.

3. Matsushime, H., L.H. Wang, and M. Shibuya, Human c-ros-1 gene homologous to the v-ros sequence of UR2 sarcoma virus encodes for a transmembrane receptorlike molecule. Mol Cell Biol, 1986. 6(8): p. 3000-4.

4. Siegel, R., D. Naishadham, and A. Jemal, Cancer statistics, 2012. CA Cancer J Clin, 2012. 62(1 ): p. 10-29.

5. Sholl, L.M., et al., ROS1 immunohistochemistry for detection of ROS1- rearranged lung adenocarcinomas. Am J Surg Pathol, 2013. 37(9): p. 1441 -9. 6. Charest, A., et al., Fusion of FIG to the receptor tyrosine kinase ROS in a glioblastoma with an interstitial del(6)(q21q21). Genes Chromosomes Cancer, 2003. 37(1 ): p. 58-71 .

7. Rimkunas, V.M., et al., Analysis of receptor tyrosine kinase ROS 1 -positive tumors in non-small cell lung cancer: identification of a FIG -ROS 1 fusion. Clin Cancer Res, 2012. 18(16): p. 4449-57.

8. Gu, T.L., et al., Survey of tyrosine kinase signaling reveals ROS kinase fusions in human cholangiocarcinoma. PLoS One, 201 1 . 6(1 ): p. e15640.

9. Birch, A.H., et al., Chromosome 3 anomalies investigated by genome wide SNP analysis of benign, low malignant potential and low grade ovarian serous tumours. PLoS One, 201 1 . 6(12): p. e28250.

10. Bergethon, K., et al., ROS1 rearrangements define a unique molecular class of lung cancers. J Clin Oncol, 2012. 30(8): p. 863-70.

1 1 . Davies, K.D., et al., Identifying and targeting ROS1 gene fusions in non -small cell lung cancer. Clin Cancer Res, 2012. 18(17): p. 4570-9.

12. Lee, J., et al., Identification of ROS1 rearrangement in gastric adenocarcinoma.

Cancer, 2013. 119(9): p. 1627-35.

13. Takeuchi, K., et al., RET, ROS1 and ALK fusions in lung cancer. Nat Med, 2012. 18(3): p. 378-81 .

14. Mok, T.S., et al., Gefitinib or carbopla tin -pa clitaxel in pulmonary adenocarcinoma. N Engl J Med, 2009. 361 (10): p. 947-57.

15. Rosell, R., et al., Erlotinib versus standard chemotherapy as first-line treatment for European patients with advanced EGFR mutation-positive non-small-cell iung cancer (EURTAC): a multicentre, open -la be I, randomised phase 3 trial. Lancet Oncol, 2012. 13(3): p. 239-46.

16. Kwak, E.L., et al., Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer. N Engl J Med, 2010. 363(18): p. 1693-703.

17. Shaw, AT., et al., Crizotinib in ROS1 -rearranged non-small-cell lung cancer. N Engl J Med, 2014. 371(21 ): p. 1963-71 .

18. Solomon, B., Validating ROS1 rearrangements as a therapeutic target in non- small-cell lung cancer. J Clin Oncol, 2015. 33(9): p. 972-4.

19. Awad, M.M., et al., Acquired resistance to crizotinib from a mutation in CD74- ROS1. N Engl J Med, 2013. 368(25): p. 2395-401 .

Song, A., et al., Molecular Changes Associated with Acquired Resistance to Cnzotinib in ROS 1 -Rearranged Non-Small Cell Lung Cancer. Clin Cancer Res, 2015. 21 (10): p. 2379-87.

Zou, H.Y., et al., PF-06463922 is a potent and selective next-generation ROS1/ALK inhibitor capable of blocking crizotinib-resistant ROS1 mutations. Proc Natl Acad Sci U S A, 2015. 112(1 1 ): p. 3493-8.

Johnson, T.W., et al., Discovery of (10R)-7-amino-12-fluoro-2, 10. 16-trimethyl- 15-OXO-10, 15, 16, 17-tetrahydro-2H-8,4-(m etheno)pyrazolo[4, 3-h][2, 5, 11]- benzoxadiazacyclotetradecine-3-carbonitrile (PF-06463922), a macrocyclic inhibitor of anaplastic lymphoma kinase (ALK) and c-ros oncogene 1 (ROS1) with preclinical brain exposure and broad-spectrum potency against ALK- resistant mutations. J Med Chem, 2014. 57(1 1 ): p. 4720-44.

Katayama, R., et al., Cabozantinib overcomes crizotinib resistance in ROS1 fusion-positive cancer. Clin Cancer Res, 2015. 21 (1 ): p. 166-74.

Davare, M.A., et al., Foretinib is a potent inhibitor of oncogenic ROS1 fusion proteins. Proc Natl Acad Sci U S A, 2013. 110(48): p. 19519-24.

Rothenstein, J.M. and N. Letarte, Managing treatment-related adverse events associated with Alk inhibitors. Curr Oncol, 2014. 21 (1 ): p. 19-26.

Ramshankar, V., et al., Molecular docking studies shows tivozanib and lapatinib as potential inhibitors of EML4-ALK translocation mediated fusion protein in non small cell lung cancer. Bioinformation, 2014. 10(10): p. 658-63.

Nakamura, K., et al., KRN951, a highly potent inhibitor of vascular endothelial growth factor receptor tyrosine kinases, has antitumor activities and affects functional vascular properties. Cancer Res, 2006. 66(18): p. 9134-42.

28. Yang, D., et al., Tivozanib reverses multidrug resistance mediated by ABCB1 (P-glycoprotein) and ABCG2 (BCRP). Future Oncol, 2014. 10(1 1 ): p. 1827-41 . 29. Shan, L, et al., Detection of R0S1 gene rearrangement in lung adenocarcinoma: comparison of IHC, FISH and real-time RT-PCR. PLoS One, 2015. 10(3): p. e0120422.

30. Grusch, M., et al., Spatio-temporally precise activation of engineered receptor tyrosine kinases by light. EMBO J, 2014. 33(15): p. 1713-26.

31 . Kim, J.M., et al., Light-dnven activation of beta 2-adrenergic receptor signaling by a chimeric rhodopsin containing the beta 2-adrenergic receptor cytoplasmic loops. Biochemistry, 2005. 44(7): p. 2284-92.

32. Ueno, H., et al., An epidermal growth factor receptor-leukocyte tyrosine kinase chimenc receptor generates ligand-dependent growth signals through the Ras signaling pathway. J. Biol. Chem., 1995. 270(34): p. 20135-42.

33. Ingles-Prieto, A., et al., Light-assisted small molecule screening against protein kinases. Nature Chemical Biology, 2015.

34. Ingles-Prieto, A., et al., O tica lly-a ctiva te d receptors I. 2013.

35. Ingles-Prieto, A., et al., Op tica lly-a ctiva te d receptors II. 2014.

36. Cravotto, G. and P. Cintas, Power ultrasound in organic synthesis: moving cavitational chemistry from academia to innovative and large-scale applications. Chem. Soc. Rev., 2006. 35(2): p. 180-96.

37. Bradley, M. and F. Grieser, Emulsion polymerization synthesis of cationic polymer latex in an ultrasonic field. J. Colloid. Interface Sci., 2002. 251 (1 ): p. 78-

84.

38. Asahi, R., et al., Visible-light photocatalysis in nitrogen-doped titanium oxides.

Science, 2001 . 293(5528): p. 269-71 .

39. Xuan, J. and W.J. Xiao, Visible-light photoredox catalysis. Angew. Chem. Int.

Ed. Engl., 2012. 51 (28): p. 6828-38.

40. Tischer, D. and O.D. Weiner, Illuminating cell signalling with optogenetic tools.

Nat. Rev. Mol. Cell. Biol., 2014. 15(8): p. 551 -8.

41 . Szobota, S. and E.Y. Isacoff, Optical control of neuronal activity. Annu. Rev.

Biophys., 2010. 39: p. 329-48.

42. Fenno, L, O. Yizhar, and K. Deisseroth, The development and application of optogenetics. Annu. Rev. Neurosci., 201 1 . 34: p. 389-412.

43. Grusch, ., et al., Spatio-temporally precise activation of engineered receptor tyrosine kinases by light. EMBO J., 2014. 33(15): p. 1713-26.

44. Mohun, T., N. Garrett, and R. Treisman, Xenopus cytoskeletal actin and human c-fos gene promoters share a conserved protein-binding site. EMBO J., 1987.

6(3): p. 667-73. Shaw, AT., et al., Tyrosine kinase gene rearrangements in epithelial malignancies. Nat Rev Cancer, 2013. 13(1 1 ): p. 772-87.

Jamil, M.O., A. Hathaway, and A. Mehta, Tivozanib: status of development. Curr Oncol Rep, 2015. 17(6): p. 24.

DiNitto, J. P., et al., Function of activation loop tyrosine phosphorylation in the mechanism of c-Kit auto-activation and its implication in sunitinib resistance. J. Biochem., 2010. 147(4): p. 601 -9.

Simard, J.R., et al., High-throughput screening to identify inhibitors which stabilize inactive kinase conformations in p38alpha. J. Am. Chem. Soc, 2009. 131 (51 ): p. 18478-88.

Burke, P., K. Schooler, and H.S. Wiley, Regulation of epidermal growth factor receptor signaling by endocytosis and intracellular trafficking. Mol. Biol. Cell., 2001 . 12(6): p. 1897-910.

Toettcher, J.E., O.D. Weiner, and W.A. Lim, Using optogenetics to interrogate the dynamic control of signal transmission by the Ras/Erk module. Cell, 2013. 155(6): p. 1422-34.

Finn, R.D., et al., The Pfam protein families database. Nucleic Acids Res., 2010. 38(Database issue): p. D21 1 -22.

Krogh, A., et al., Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J. Mol. Biol., 2001 . 305(3): p. 567-80.

Edgar, R.C., Search and clustering orders of magnitude faster than BLAST. Bioinformatics, 2010. 26(19): p. 2460-1 .

Engler, C. and S. Marillonnet, Golden Gate cloning. Methods Mol Biol, 2014. 1116: p. 1 19-31 .

Chen, Z., et al., Non-small-cell lung cancers: a heterogeneous set of diseases. Nat Rev Cancer, 2014. 14(8): p. 535-46.

Finn, R.D., et al., Pfam: the protein families database. Nucleic Acids Res., 2014. 42(Database issue): p. D222-30.