WO2022053708A1

WO2022053708A1 - Pharmaceutical combinations for use in the treatment of gastric adenocarcinoma

Info

Publication number: WO2022053708A1
Application number: PCT/EP2021/075216
Authority: WO
Inventors: Paul Mcsheehy; Mahmoud EL SHEMERLY
Original assignee: Basilea Pharmaceutica International AG
Priority date: 2020-09-14
Filing date: 2021-09-14
Publication date: 2022-03-17

Abstract

The present invention provides pharmaceutical combinations comprising derazantinib and paclitaxel (1) for use in the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene; and (2) for use in the treatment of a patient with a solid tumor, wherein the solid tumor has a level of M2-tumor associated macrophages determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics: (i) the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene; (ii) the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value; (iii) the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

Description

Pharmaceutical Combinations for Use in the Treatment of Gastric Adenocarcinoma

The present invention relates to methods of using pharmaceutical combinations as described herein in the treatment of gastric adenocarcinoma.

Gastric adenocarcinoma is the most common histological type (-95%) of all stomach malignancies and can originate from the fundus, corpus or antrum of the stomach. Gastro-esophageal adenocarcinomas originate from the cardia and gastro-esophageal junction region and are further anatomically classified by the Siewert-Stein classification (Siewert et al. Br J Surg 1998;85: 1457-1459). Gastric and gastroesophageal adenocarcinomas (hereafter sub-summarized under ‘gastric adenocarcinoma’ and abbreviated as GAC) have two major histological phenotypes: the intestinal type with well-to-moderately- well- differentiated histology and the diffuse type with poorly differentiated histology (Ajani et al. Nat. Rev. Dis. Primers 2017;3: 17036; Lauren et al. Acta Pathol. Microbiol. Scand. 1965;64, 31-49). A World Health Organization (WHO) classification provides detailed descriptions of histological features and classifies GAC into tubular, papillary, mucinous, mixed and poorly cohesive carcinomas (Bosman et el. WHO Classification of Tumors of the Digestive System 4th edn Vol. 48-58 (IARC, 2010).3).

In 2018, there were an estimated 1,033,700 new cases of GAC and 783,000 deaths from the disease worldwide (Bray et al., CA Cancer J Clin. 2018;68:394-424), making it the fifth most frequently diagnosed cancer and the third leading cause of cancer death. Incidences are 2-fold higher in men than in women, with rates of 32.1 and 13.2 per 100,000 men and women, respectively, in Eastern Asia (e.g. in Mongolia, Japan and the Republic of Korea (the country with the highest rates worldwide in both sexes)). The incidence in Northern America is generally low with rates of 5.6 and 2.8 per 100,000 men and women, respectively (Bray et al.). There is a notable difference across Europe with incidence rates lowest in Northern Europe (6.2 and 3.1 per 100,000 men and women, respectively) increasing to 8.2/3.7, 10.4/5.0 and 17.1/7.5 per 100,000 men and women in Western, Southern and Eastern Europe, respectively (Bray et al.). While early stage GAC is associated with a 5-year survival rate of up to 95% (Crew 2006 World J. Gastroenterol. 2006; 12:354-362), advanced-stage GAC (which cannot be surgically treated) has a median survival of -9-10 months (Ajani et al. J. Natl Compr. Cane. Netw. 2016; 14: 1286-1312). Despite improvements in endoscopic, surgical and systemic treatments the global 5-year survival rates remain unsatisfactory (-25-30%; Verdecchia et al. Lancet Oncol 2007;8:784-796), with the exception of those in Japan and South Korea (> 50%; Cancer Epidemiol Biomarkers Prev 2016;25 : 16-27), a difference that can be attributed to early detection efforts in these Asian countries.

Fibroblast Growth Factor Receptor (FGFR) genes are a family of four genes (FGFR1, FGFR2, FGFR3 and FGFR4) which have a role in several biological processes, including regulation of development and tissue repair. FGFR proteins function as transmembrane receptors for FGF family members. They are receptor tyrosine-kinases and include an extracellular ligand-binding domain encompassing three immunoglobulin (Ig)-like domains (Igl, Igll and Iglll), a transmembrane domain and an intracellular tyrosine-kinase domain region encompassing two tyrosine kinase subdomains (TKI and TKII). They are expressed in a number of alternatively spliced isoforms. Binding of an FGF ligand induces receptor dimerization and trans-autophosphorylation of tyrosine residues in the cytoplasmic kinase domain. This activation of the protein tyrosine-kinase domain triggers the recruitment of downstream signaling molecules which results in the activation of various signaling cascades. Missense mutations, chromosomal translocations or aberrant splicing in the FGFR can result in receptor activation through a variety of mechanisms. Constitutive activation of FGFR promotes tumor growth via activation of downstream signaling cascades (Katoh et al. Nature Reviews Clinical Oncology 2019; 16(2): 105— 122; Babina et al. Nature Reviews Cancer 2017, 17:318-332).

Derazantinib is an investigational orally administered small-molecule FGFR kinase inhibitor with strong activity against FGFR1, FGFR2, and FGFR3 (Hall et al. PLoS One. 2016; 1 l(9):e0162594) currently in clinical trials. It has the chemical structure shown below:

Paclitaxel is a well-known small molecule approved for the treatment of a number of cancer indications, including advanced gastric or gastroesophageal junction (GEJ) adenocarcinoma in combination with ramuciramab (Wilke et al. Lancet Oncol 2014;15: 1224-1235). It is commercialized underthe brand name Taxol®.

There is an ongoing need for new effective treatment options for GAC patients. As demonstrated in the Examples below, it has now surprisingly been found that combinations of the two compounds described above provide positive outcomes in gastric cancer models.

Summary of the invention

In a first aspect, the invention provides a pharmaceutical combination comprising derazantinib and paclitaxel for use in the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene. In a further aspect, the invention provides derazantinib for use in combination with paclitaxel for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

In a further aspect, the invention provides paclitaxel for use in combination with derazantinib for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

In a further aspect, the invention provides use of derazantinib and paclitaxel in the manufacture of single agent medicaments for use in combination for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

In a further aspect, the invention provides use of derazantinib in the manufacture of a medicament for use in combination with paclitaxel for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

In a further aspect, the invention provides use of paclitaxel in the manufacture of a medicament for use in combination with derazantinib for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

In a further aspect, the invention provides a method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of derazantinib and a therapeutically effective amount of paclitaxel, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

In a further aspect, the invention provides a method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of derazantinib, which patient is undergoing or will undergo treatment with paclitaxel, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

In a further aspect, the invention provides a method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of paclitaxel, which patient is undergoing or will undergo treatment with derazantinib, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

Further aspects and embodiments of the invention relate to the finding that higher levels of M2 tumor associated macrophages (TAMs) are correlated with good outcomes when the two compounds above are combined in gastric cancer models. Accordingly the gastric adenocarcinoma in each of the above aspects of the invention may have a level of TAMs that is higher than a standard value.

In a further aspect, the invention provides a pharmaceutical combination comprising derazantinib and paclitaxel for use in the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics :

(i) the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene;

(ii) the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value;

(iii) the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

In a further aspect, the invention provides derazantinib for use in combination with paclitaxel for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics :

In a further aspect, the invention provides paclitaxel for use in combination with derazantinib for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics :

(iii) the solid tumor has a level of mRNA expression of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value. In a further aspect, the invention provides use of derazantinib and paclitaxel in the manufacture of single agent medicaments for use in combination for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

In a further aspect, the invention provides use of derazantinib in the manufacture of a medicament for use in combination with paclitaxel for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

In a further aspect, the invention provides use of paclitaxel in the manufacture of a medicament for use in combination with derazantinib for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

(iii) the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value. In a further aspect, the invention provides a method of treating a patient with a solid tumor (e.g. gastric adenocarcinoma) comprising administering to the patient a therapeutically effective amount of derazantinib and a therapeutically effective amount of paclitaxel, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

In a further aspect, the invention provides a method of treating a patient with a solid tumor (e.g. gastric adenocarcinoma) comprising administering to the patient a therapeutically effective amount of derazantinib, which patient is undergoing or will undergo treatment with paclitaxel, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the gastric adenocarcinoma has at least one of the following characteristics:

In a further aspect, the invention provides a method of treating a patient with a solid tumor (e.g. gastric adenocarcinoma) comprising administering to the patient a therapeutically effective amount of paclitaxel, which patient is undergoing or will undergo treatment with derazantinib, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the gastric adenocarcinoma has at least one of the following characteristics:

(iii) the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value. The solid tumor (e.g. gastric adenocarcinoma) may harbor only one mutated FGFR gene, e.g. a mutated FGFR1 gene or a mutated FGFR2 gene, or a mutated FGFR3 gene, or the gastric adenocarcinoma may harbor more than one mutated FGFR gene, e.g. a mutated FGFR1 gene and an FGFR2 gene, e.g. a mutated FGFR2 gene and a mutated FGFR3 gene, e.g. a mutated FGFR1 and an FGFR3 gene, e.g. a mutated FGFR1 gene, a mutated FGFR2 gene and a mutated FGFR3 gene.

Additional aspects and embodiments of the invention are described in more detail below.

Brief Description of Figures:

Figure 1: Figure 1 shows the efficacy (1A) and tolerability (IB) of the combination of derazantinib and paclitaxel in the GA3055 PDX-model. The FGFR genetic -aberration in this tumor model is an FGFR2- fusion (LINCO1153).

Figure 2: Figure 2 shows the efficacy (2A, 2C) and tolerability (2B, 2D) of the combination of derazantinib and paclitaxel in the GA6208 PDX-model. This model has two different FGFR genetic- aberrations, an FGFR2 -mutation (Y244C) and an FGFR3 mutation (G224S).

Figure 3 : Figure 3 shows the efficacy (3 A) and tolerability (3B) of the combination of derazantinib and paclitaxel in the GA3236 PDX-model. This is an FGFR wild type tumor model with moderately-high expression of FGFR3 (RNA-Seq Log2=6.9).

Figure 4: Figure 4 shows the efficacy (3 A) and tolerability (3B) of the combination of derazantinib and paclitaxel in the GA0095 PDX-model. This is an FGFR wild type tumor model with moderately-high expression of FGFR3 (RNA-Seq Log2=7.3).

Figure 5 : Figure 5 shows the efficacy (5 A) and tolerability (5B) of the combination of derazantinib and paclitaxel in the GAO 114 PDX-model. This tumor model has high amplification of FGFR2.

Figure 6: Figure 6 shows the efficacy (6A, 6C) and tolerability (6B, 6D) of the combination of derazantinib and paclitaxel in the SNU-16 CDX-model. The FGFR genetic-aberration in this tumor model is an FGFR2 -fusion (PDHX).

Figure 7 : Figure 7 shows the efficacy (7A) and tolerability (6B) of the combination of derazantinib and paclitaxel in the GA0031 PDX-model. The FGFR genetic -aberration in this tumor model is an FGFR1- mutation (R285W (R254W with respect to SEQ ID NO: 2)). Figure 8: Figure 8 shows a plot of CCI Index value versus median %M2 -Tumor associated macrophages. The dots represent the values for the PDX and CDX models (the individual data points are shown in Table 3).

Figure 9: Figure 9 shows the FGFR1 canonical DNA sequence (9A, SEQ ID NO: 1) and canonical amino acid sequence (9B, SEQ ID NO: 2). Exons are located as follows: exon 1: 1 to 655; exon 2: 656 to 834; exon 3: 835 to 1101; exon 4: 1102 to 1191; exon 5: 1192 to 1364; exon 6: 1365 to 1488; exon 7: 1489 to 1679; exon 8: 1680 to 1824; exon 9: 1825 to 2027; exon 10: 2028 to 2173; exon 11: 2174 to 2295; exon 12: 2296 to 2406; exon 13: 2407 to 2597; exon 14: 2598 to 2720; exon 15: 2721 to 2791; exon 16: 2792 to 2929; exon 17: 2930 to 3035; exon 18: 3036 to 5697. The signal peptide is encoded by 744 to 806. The sequences can be found under Genebank accession no. NM_023110.

Figure 10: Figure 10 shows the FGFR2 canonical DNA sequence (10A, SEQ ID NO: 3) and canonical amino acid sequence (10B, SEQ ID NO: 4). Exons are located as follows: exon 1: 1 to 483; exon 2: 484 to 742; exon 3: 743 to 1009; exon 4: 1010 to 1087; exon 5: 1088 to 1257; exon 6: 1258 to 1381; exon 7: 1382 to 1572; exon 8: 1573 to 1717; exon 9: 1718 to 1920; exon 10: 1921 to 2072; exon 11: 2073 to 2194; exon 12: 2195 to 2305; exon 13: 2306 to 2496; exon 14: 2497 to 2619; exon 15: 2620 to 2690; exon 16: 2691 to 2828; exon 17: 2829 to 2934; exon 18: 2935 to 4624. The signal peptide is encoded by 634-696. The sequences can be found under Genebank accession no. NM_000141.

Figure 11: Figure 11 shows the FGFR3 canonical DNA sequence (11A, SEQ ID NO: 5) and canonical amino acid sequence (1 IB, SEQ ID NO: 6). Exons are located as follows: exon 1: 1 to 173; exon 2: 174- 384; exon 3: 385 to 654; exon 4: 655 to 720; exon 5: 721 to 890; exon 6: 891 to 1014; exon 7: 1015 to 1205; exon 8: 1206 to 1350; exon 9: 1351 to 1541; exon 10: 1542 to 1687; exon 11: 1688 to 1809; exon 12: 1810 to 1920; exon 13: 1921 to 2111; exon 14: 2112 to 2234; exon 15: 2235 to 2305; exon 16: 2306 to 2443; exon 17: 2444 to 2549; exon 18: 2550 to 4301. The signal peptide is encoded by 276 to 341. The sequences can be found under Genebank accession no. NM_000142.

Figure 12: Figure 12 shows a FGFR1 common isoform DNA sequence (12A, SEQ ID NO: 7) and corresponding amino acid sequence (12B, SEQ ID NO: 8). Exons are located as follows: exon 1: 1 to 655; exon 2: 656 to 834; exon 3: 835 to 1101; exon 4: 1102 to 1185; exon 5: 1186 to 1358; exon 6: 1359 to 1482; exon 7: 1483 to 1673; exon 8: 1674 to 1818; exon 9: 1819 to 2021; exon 10: 2022 to 2167; exon 11: 2168 to 2289; exon 12: 2290 to 2400; exon 13: 2401 to 2591; exon 14: 2592 to 2714; exon 15: 2715 to 2785; exon 16: 2786 to 2923; exon 17: 2924 to 3029; exon 18: 3030 to 5691. The signal peptide is encoded by 744-806. The sequences can be found under Genebank accession no. NM_015850.4. Figure 13: Figure 13 shows a FGFR2 common isoform DNA sequence (13A, SEQ ID NO: 9) and corresponding amino acid sequence (13B, SEQ ID NO: 10). Exons are located as follows: exon 1: 1 to 497; exon 2: 498 to 756; exon 3: 757 to 1023; exon 4: 1024 to 1101; exon 5: 1102 to 1271; exon 6: 1272 to 1395; exon 7: 1396 to 1586; exon 8: 1587 to 1734; exon 9: 1735 to 1937; exon 10: 1938 to 2089; exon 11:2090 to 2211; exon 12: 2212 to 2322; exon 13: 2323 to 2513; exon 14: 2514 to 2636; exon 15: 2637 to 2707; exon 16: 2708 to 2845; exon 17: 2846 to 2951; exon 18: 2952 to 4643. The signal peptide is encoded by 648-710. The sequences can be found under Genebank accession no. NM_022970.3.

Figure 14: Figure 14 shows a FGFR3 common isoform DNA sequence (14A, SEQ ID NO: 11) and corresponding amino acid sequence (14B, SEQ ID NO: 12). Exons are located as follows: exon 1: 1 to 173; exon 2: 174 to 384; exon 3: 385 to 654; exon 4: 655 to 720; exon 5: 721 to 890; exon 6: 891 to 1014; exon 7: 1015 to 1205; exon 8: 1206 to 1356; exon 9: 1357 to 1547; exon 10: 1548 to 1693; exon 11: 1694 to 1815; exon 12: 1816 to 1926; exon 13: 1927 to 2117; exon 14: 2118 to 2240; exon 15: 2241 to 2311; exon 16: 2312 to 2449; exon 17: 2450 to 2555; exon 18: 2556 to 4307. The signal peptide is encoded by 276-341. The sequences can be found under Genebank accession no. NM_001163213.2.

Detailed description of the invention

Definitions

Certain terms used herein are described below. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as is commonly understood by one of skill in the art to which this invention belongs.

The term "combination," "therapeutic combination," or "pharmaceutical combination" are interchangeable terms and refer to either a fixed combination in one dosage unit form, or a non-fixed combination in which each active principle is formulated as a separate dosage unit form, or a kit, e.g. a kit of parts, for the combined administration, where two or more therapeutic agents may be administered independently, at the same time or separately within time intervals, especially where these time intervals allow that the combination partners show a cooperative, e.g. synergistic, effect. Usually the active principles are provided as separate dosage forms for independent administration.

The term "combination therapy" refers to the administration of two or more therapeutic agents to treat a therapeutic condition or disorder described in the present disclosure. Such administration encompasses co-administration of these therapeutic agents in a substantially simultaneous manner as well as use of each type of therapeutic agent in a sequential and/or separate manner (e.g. according to different administration routes), either at approximately the same time or at different times, e.g. according to different dosage regimens. When the therapeutic agents are administered sequentially and/or separately the dosing schedules will be such that there is a therapeutic interaction between the therapeutic agents within the patient’s body and/or that a therapeutic effect resulting from the first therapeutic agent is present when the second therapeutic agent is administered. For example, when the agents are administered according to cyclic treatment schedules, the cyclic treatment schedules may overlap, or when one therapeutic agent is administered according to a continuous dosing schedule and the second according to a cyclic schedule, then at least one dose from the agent administered according to the continuous schedule will occur during the treatment cycle of the other therapeutic agent. Usually there will be at least one interval of no more than seven days between doses of the two therapeutic agents when administered according to a cyclic treatment schedule.

The term "pharmaceutical composition" is defined herein to refer to a solid or liquid formulation containing at least one therapeutic agent to be administered to a patient, optionally with one or more pharmaceutically acceptable excipients, in order treat a particular disease or condition affecting the patient.

The term "pharmaceutically acceptable" as used herein refers to items such as compounds, materials, compositions and/or dosage forms, which are, within the scope of sound medical judgment, suitable for contact with the tissues of a human, without excessive toxicity or other complications commensurate with a reasonable benefit/risk ratio.

The terms "fixed combination", "fixed dose" and "single formulation" as used herein refers to a single carrier or vehicle or dosage form formulated to deliver an amount, which is jointly therapeutically effective for the treatment of neoplastic diseases, of both therapeutic agents to a patient. The single vehicle is designed to deliver an amount of each of the agents, along with any pharmaceutically acceptable carriers or excipients.

The term "non-fixed combination,", “kit”, and "separate formulations" means that the active ingredients, are both administered to a patient as separate entities either simultaneously, concurrently or sequentially, wherein such administration provides therapeutically effective levels of the two compounds in the body of the patient.

The term “patient” refers to a human presenting themselves for therapeutic treatment.

The term "treatment," as used herein in the context of treating a disease in a patient pertains generally to treatment and therapy in which some desired therapeutic effect is achieved, for example one or more of the following: the inhibition of the progress of the disease, a reduction in the rate of progress, a halt in the rate of progress, a prevention of the progression of the disease, alleviation of symptoms of the disease, amelioration of disease, and cure of the disease. For example, treatment can be the diminishment of one or several symptoms of a disorder or complete eradication of a disorder. Within the meaning of the present disclosure, the term "treat" also denotes to arrest, delay the onset (i.e., the period prior to clinical manifestation of a disease) and/or reduce the risk of developing or worsening a disease.

The term "prevent", "preventing" or "prevention" as used herein comprises the prevention of at least one symptom associated with or caused by disease being prevented.

The term "pharmaceutically effective amount," "therapeutically effective amount," or "clinically effective amount" of a combination of therapeutic agents is an amount sufficient to provide an observable or clinically significant improvement over the baseline clinically observable signs and symptoms of the disease treated with the combination, e.g. commensurate with a reasonable benefit/risk ratio, when administered in accordance with a desired treatment regimen. The skilled person will understand that the therapeutically effective amount of an agent for use in combination therapy may be lower than the amount required to provide a therapeutic effect when using the agent as a monotherapy.

The term “about” means a variation of no more than 10% of the relevant figure, preferably no more than J CO //o.

The term “gastric adenocarcinoma” includes gastric adenocarcinomas and gastro-esophageal adenocarcinomas .

The term “stratifying patients” means identifying a patient or group of patients as belonging to a group of patients who are likely to benefit from receiving a particular treatment.

For convenience, reference to derazantinib refers to the compound, e.g. in free form, and pharmaceutically acceptable salts thereof, such as hydrochloride salts, in particular the dihydrochloride salt. In addition, reference to derazantinib includes all possible solvates and complexes (including hydrates) of derazantinib as well as any polymorphs, including amorphous solids, as well as pharmaceutically acceptable salts of any of the foregoing, including those described in WO 2017/106642 and WO 2017/106639. The same applies to paclitaxel, although paclitaxel is usually used in free form.

Mutated FGFR genes

For simplification, the coding regions of the FGFR genes are represented by the so-called “canonical transcripts”. These canonical forms depict the DNA sequence of exons without taking into account the alternative splicing. The FGFR1 gene canonical form and corresponding amino acid canonical form of the most common FGFR1 isoform which is generally used in the field as the reference FGFR1 sequence is represented by SEQ ID NOs: 1 and 2 respectively. The FGFR2 gene canonical form of the most common FGFR2 isoform which is generally used in the field as the reference FGFR2 sequence and corresponding amino acid canonical form is represented by SEQ ID NOs: 3 and 4 respectively. The FGFR3 gene canonical form of the most common FGFR3 isoform which is generally used in the field as the reference FGFR3 sequence and corresponding amino acid canonical form is represented by SEQ ID NOs: 5 and 6 respectively. Exons 11-17 encode the tyrosine kinase domain in each case. Reference herein to the tyrosine kinase domain refers to the region of the gene encoding both TKI and TKII.

The FGFR1 gene canonical form and corresponding amino acid canonical form of a second common FGFR1 isoform is represented by SEQ ID NOs: 7 and 8 respectively. The FGFR2 gene canonical form of a second common FGFR2 isoform and corresponding amino acid canonical form is represented by SEQ ID NOs: 9 and 10 respectively. The FGFR3 gene canonical form of a second common FGFR3 isoform and corresponding amino acid canonical form is represented by SEQ ID NOs: 11 and 12 respectively.

Reference herein to a mutated FGFR1 gene refers to an FGFR1 gene which encodes an amino acid sequence that comprises at least one change compared to the amino acid sequence encoded by a reference FGFR1 DNA sequence. Reference herein to a mutated FGFR2 gene refers to an FGFR2 gene which encodes an amino acid sequence that comprises at least one change compared to the amino acid sequence encoded by a reference FGFR2 DNA sequence. Reference herein to a mutated FGFR3 gene refers to an FGFR3 gene which encodes an amino acid sequence that comprises at least one change compared to the amino acid sequence encoded by a reference FGFR3 DNA sequence. A change in the amino acid sequence may be an insertion, a deletion, a substitution and/or an addition (e.g. to the N or C terminus) of at least one (including more than one) amino acid.

In some embodiments, the reference FGFR1 DNA sequence may be the corresponding wildtype DNA sequence, i.e. the FGFR1 DNA sequence from matched normal (non-tumor) tissue from the same subject. In some embodiments, the reference FGFR2 DNA sequence may be the corresponding wildtype DNA sequence, i.e. the FGFR2 DNA sequence from matched normal (non-tumor) tissue from the same subject. In some embodiments, the reference FGFR3 DNA sequence may be the corresponding wildtype DNA sequence, i.e. the FGFR3 DNA sequence from matched normal (non-tumor) tissue from the same subject. A comparison with the wildtype DNA sequence will take into account any germline variants or somatic mutations present in the DNA sequence from normal tissue. The at least one change in the amino acid sequence of the FGFR protein encoded by the tumor FGFR DNA sequence is preferably not a germline variant or a somatic mutation present in the DNA sequence from normal tissue.

A drawback of directly comparing tumor DNA against DNA from matched normal tissue is that double the number of samples must be obtained, sequenced and analyzed. Other widely-used approaches to identify tumor mutations without using matched normal samples rely instead on DNA sequence databases, such as the dbSNP database (htps ://www. ncbi .nlm. nih.gov/snD/) or the Genome Analysis ToolKit (htps ://gatk. broadinstitute . org/hc) to eliminate germline variations and somatic mutations not associated with the tumor. See Teer et al. Human Genomics 2017; 11:22 and Duo et al. Nat Biotechnol. 2020; 38(3): 314-319.

In some embodiments, the reference FGFR1 DNA sequence may be a database of multiple FGFR1 DNA sequences in which sequences with known germline variations and somatic mutations from normal tissue are taken into account and not identified (“called”) as a mutation (i.e. not identified as a tumor-associated mutation). In some embodiments, the reference FGFR2 DNA sequence may be a database of multiple FGFR2 DNA sequences in which sequences with known germline variations and somatic mutations from normal tissue are taken into account and not identified (“called”) as a mutation (i.e. not identified as a tumor-associated mutation). In some embodiments, the reference FGFR3 DNA sequence may be a database of multiple FGFR3 DNA sequences in which sequences with known germline variations and somatic mutations from normal tissue are taken into account and not identified (“called”) as a mutation (i.e. not identified as a tumor-associated mutation). Such a database may be a public variation database or may be a “panel of normals” consisting of the DNA sequences of healthy unrelated individuals, e.g. 400 or more individuals. Use of a combination of both a public variation database and a panel of normals may allow higher accuracy than just using one of these methods, i.e. in the elimination of germline variants and non-tumor somatic mutations.

In some embodiments, for simplicity the reference FGFR amino acid sequences may be represented by the canonical form of the respective amino acid sequences. Accordingly the reference amino acid sequence of FGFR1 may be represented by SEQ ID NO: 2 or SEQ ID NO: 8, the reference amino acid sequence of FGFR2 may be represented by SEQ ID NO: 4 or SEQ ID NO: 10, the reference amino acid sequence of FGFR3 may be represented by SEQ ID NO: 6 or SEQ ID NO: 12. In some embodiments the reference amino acid sequence of FGFR1 may be represented by SEQ ID NO:2. In some embodiments the reference amino acid sequence of FGFR1 may be represented by SEQ ID NO: 8. In some embodiments the reference amino acid sequence of FGFR2 may be represented by SEQ ID NO:4. In some embodiments the reference amino acid sequence of FGFR2 may be represented by SEQ ID NO: 10. In some embodiments the reference amino acid sequence of FGFR3 may be represented by SEQ ID NO:6. In some embodiments the reference amino acid sequence of FGFR3 may be represented by SEQ ID NO: 12. In some embodiments the reference amino acid sequence of FGFR1 may be represented by SEQ ID NO:2, the reference amino acid sequence of FGFR2 may be represented by SEQ ID NO: 4, the reference amino acid sequence of FGFR3 may be represented by SEQ ID NO: 6.

In some embodiments, the mutated FGFR1 gene encodes a functional FGFR1 tyrosine -kinase domain, e.g. which has phosphorylation activity. For example, the protein encoded by the mutant FGFR1 gene may retain at least 20% of the phosphorylation activity of the FGFR1 protein encoded by the wildtype FGFR1 gene, e.g. at least 30%, e.g. at least 40%, e.g. at least 50%, e.g. at least 60%, e.g. at least 70%, e.g. at least 80%, e.g. at least 90%, e.g. at least 95%, e.g. at least 99%, e.g. at least 100%, e.g. more than 100% of the phosphorylation activity of the protein encoded by the FGFR1 wildtype gene.

In some embodiments, the mutated FGFR2 gene encodes a functional FGFR2 tyrosine -kinase domain, e.g. which has phosphorylation activity. For example, the protein encoded by the mutant FGFR2 gene may retain at least 20% of the phosphorylation activity of the FGFR2 protein encoded by the wildtype FGFR2 gene, e.g. at least 30%, e.g. at least 40%, e.g. at least 50%, e.g. at least 60%, e.g. at least 70%, e.g. at least 80%, e.g. at least 90%, e.g. at least 95%, e.g. at least 99%, e.g. at least 100%, e.g. more than 100% of the phosphorylation activity of the protein encoded by the FGFR2 wildtype gene.

In some embodiments, the mutated FGFR3 gene encodes a functional FGFR3 tyrosine -kinase domain, e.g. which has phosphorylation activity. For example, the protein encoded by the mutant FGFR3 gene may retain at least 20% of the phosphorylation activity of the FGFR3 protein encoded by the wildtype FGFR3 gene, e.g. at least 30%, e.g. at least 40%, e.g. at least 50%, e.g. at least 60%, e.g. at least 70%, e.g. at least 80%, e.g. at least 90%, e.g. at least 95%, e.g. at least 99%, e.g. at least 100%, e.g. more than 100% of the phosphorylation activity of the protein encoded by the FGFR3 wildtype gene.

The level of phosphorylation activity may be determined by recombinantly expressing and isolating a protein fragment containing the tyrosine kinase domain from Sf9 insect or E. coli cells, and the phosphorylation activity may be determined in vitro by using a protein kinase assay, e.g. the ³³PanQinase® Activity Assay, performed as follows: a 50 pl reaction volume contains 25 pl of assay buffer (70 mM HEPES-NaOHpH 7.5, 3 mM MgCh, 3 mM MnCT. 3 pM Na-orthovanadate, 1.2 mM DTT, 50 pg/ml PEG20000, ATP (corresponding to the apparent ATP-Km of the kinase, [y-³³P]-ATP), 5 pl of ATP solution, 20 pl enzyme/substrate mix. The reaction cocktails are incubated at 30°C for 60 minutes. The reaction is stopped with 50 pl of 2 % (v/v) H3PO4, plates are aspirated and washed two times with 200 pl 0.9 % (w/v) NaCl. Incorporation of ³³P is determined with a microplate scintillation counter. The protein fragment may correspond to any amino acid sequence derived from FGFR2, FGFR1 and FGFR3, provided that fragment contains the tyrosine-kinase domain.

In some embodiments, the mutated FGFR1 gene encodes an FGFR1 protein comprising a functional kinase domain which is constitutively active, e.g. it is able to self-phosphorylate and bind and/or phosphorylate downstream signaling molecules. The encoded mutant FGFR1 protein may exhibit kinase activity in the absence of the respective FGF ligand. The mutation may cause the FGFR1 gene to be a onco-driver gene, e.g. a gene that gives cells a growth advantage when mutated, thereby facilitating tumor proliferation. In some embodiments, the mutated FGFR2 gene encodes an FGFR2 protein comprising a functional kinase domain which is constitutively active, e.g. it is able to self-phosphorylate and bind and/or phosphorylate downstream signaling molecules. The encoded mutant FGFR2 protein may exhibit kinase activity in the absence of the respective FGF ligand. The mutation may cause the FGFR2 gene to be an onco-driver gene, e.g. a gene that gives cells a growth advantage when mutated, thereby facilitating tumor proliferation.

In some embodiments, the mutated FGFR3 gene encodes an FGFR3 protein comprising a functional kinase domain which is constitutively active, e.g. it is able to self-phosphorylate and bind and/or phosphorylate downstream signaling molecules. The encoded mutant FGFR3 protein may exhibit kinase activity in the absence of the respective FGF ligand. The mutation may cause the FGFR3 gene to be an onco-driver gene, e.g. a gene that gives cells a growth advantage when mutated, thereby facilitating tumor proliferation.

A tyrosine kinase domain may retain functionality despite the presence of amino acid mutations in the tyrosine kinase domain. Nevertheless, for simplicity in some embodiments, a mutated FGFR1 gene may refer to an FGFR1 gene which encodes a non-mutated tyrosine kinase domain, namely amino acids 480 to 757 of SEQ ID NO: 2. In this case the mutation may be at least one change (insertion, deletion, substitution and/or addition of at least one amino acid) in the amino acid sequence outside of this region. Likewise, in some embodiments, a mutated FGFR2 gene may refer to an FGFR2 gene encoding a nonmutated tyrosine kinase domain, namely amino acids 480 to 760 of SEQ ID NO: 4. In this case the mutation may be at least one change (insertion, deletion, substitution and/or addition of at least one amino acid) in the amino acid sequence outside of this region. Similarly, in some embodiments, a mutated FGFR3 gene may refer to an FGFR3 gene encoding a non-mutated tyrosine kinase domain, namely amino acids 471 to 751 of SEQ ID NO: 6. In this case the mutation may be at least one change (insertion, deletion, substitution and/or addition of at least one amino acid) in the amino acid sequence outside of this region.

Some tumor FGFR genes may code for a protein which has reduced or even no kinase activity. This may be caused for example by a partial deletion, a point mutation or an insertion in the tyrosine kinase domain - although not all such mutations will cause a loss of kinase activity.

In some embodiments, FGFR1 genes encoding FGFR1 proteins with reduced or no phosphorylation activity may be excluded as mutated FGFR1 genes for the purposes of this invention. For example, the phosphorylation activity of the such proteins may have no more than 50%, e.g. no more than 40%, e.g. no more than 30% e.g. no more than 20%, e.g. no more than 10%, e.g. no more than 5%, e.g. 0% of the phosphorylation activity of the protein encoded by the corresponding wildtype FGFR1 gene. Phosphorylation activity may be determined in vitro using a protein kinase assay as described above.

In some embodiments, FGFR1 genes that harbor the following may be excluded as mutated FGFR1 genes for the purposes of this invention: i) a nonsense and/or frameshift mutations in exons 2-16 and in exon 17 up to amino acid 757 (SEQ ID NO: 2); ii) a large deletion in exons 11-17, e.g. deletion of a region encoding at least 30 consecutive amino acids e.g. at least 50 consecutive amino acids; iii) a DNA sequence encoding a point amino acid substitution which inactivates tyrosine kinase activity; iv) splice variants in introns 10-15, e.g. splice variants which result in missing exons from the region encoding the tyrosine kinase domain; and v) a 5’ FGFR1 fusion with a breakpoint in introns 2-9.

In some embodiments, FGFR2 genes encoding FGFR2 proteins with reduced phosphorylation activity may be excluded as mutated FGFR2 genes for the purposes of this invention. For example, the phosphorylation activity of the such proteins may have no more than 50%, e.g. no more than 40%, e.g. no more than 30% e.g. no more than 20%, e.g. no more than 10%, e.g. no more than 5%, e.g. 0% of the phosphorylation activity of the protein encoded by the corresponding wildtype FGFR2 gene. Phosphorylation activity may be determined in vitro using a protein kinase assay as described above.

In some embodiments, FGFR2 genes that harbor the following may be excluded as mutated FGFR2 genes for the purposes of this invention: i) a nonsense and/or frameshift mutations in exons 2-16 and in exon 17 up to amino acid 761 (SEQ ID NO: 4); ii) a large deletion in exons 11-17, e.g. deletion of a region encoding at least 30 consecutive amino acids e.g. at least 50 consecutive amino acids; iii) a DNA sequence encoding a point amino acid substitution which inactivates tyrosine kinase activity; iv) splice variants in introns 10-15, e.g. splice variants which result in missing exons from the region encoding the tyrosine kinase domain; and v) a 5’ FGFR2 fusion with a breakpoint in introns 2-9. In some embodiments, FGFR3 genes encoding FGFR3 proteins with reduced phosphorylation activity may be excluded as mutated FGFR3 genes for the purposes of this invention. For example, the phosphorylation activity of the such proteins may have no more than 50%, e.g. no more than 40%, e.g. no more than 30% e.g. no more than 20%, e.g. no more than 10%, e.g. no more than 5%, e.g. 0% of the phosphorylation activity of the protein encoded by the corresponding wildtype FGFR3 gene. Phosphorylation activity may be determined in vitro using a protein kinase assay as described above.

In some embodiments, FGFR3 genes that harbor the following may be excluded as mutated FGFR3 genes for the purposes of this invention: i) a nonsense and/or frameshift mutations in exons 2-16 and in exon 17 up to amino acid 751 (SEQ ID NO: 6); ii) a large deletion in exons 11-17, e.g. deletion of a region encoding at least 30 consecutive amino acids e.g. at least 50 consecutive amino acids iii) a DNA sequence encoding a point amino acid substitution which inactivates tyrosine kinase activity; iv) splice variants in introns 10-15, e.g. splice variants which result in missing exons from the region encoding the tyrosine kinase domain. and v) a 5’ FGFR3 fusion with a breakpoint in introns 2-9.

In some embodiments, a mutated FGFR1 gene may be a gene rearrangement, a splice site variant, a short variant or a combination of the foregoing. In some embodiments, a mutated FGFR2 gene may be a gene rearrangement, a splice site variant, a short variant or a combination of the foregoing. In some embodiments, a mutated FGFR3 gene may be a gene rearrangement, a splice site variant, a short variant or a combination of the foregoing. Gene rearrangements represent larger changes in sequence, often with larger pieces (e.g. more than 200 consecutive base pairs) of DNA translocated, inserted or deleted compared to the original DNA sequence of a gene. Splice site variants are alterations in the DNA sequence that occur at the boundary of an exon and an intron (the splice site) and can result in changes to the open reading frame. Short variants represent insertion or deletion of smaller nuclear variants (e.g. up to 200 consecutive base pairs).

An FGFR gene rearrangement may be any type of gene rearrangement arising for example as a result of a chromosomal translocation. Usually gene rearrangement events are caused by a breakage in the DNA double helices at two different locations, followed by a re-joining of the broken ends to produce a new arrangement of genes and intergenetic sequences on the chromosome, different from the gene order of the chromosomes before they were broken. A fusion gene is an example of a gene rearrangement. An FGFR fusion gene is a hybrid gene formed from an or parts of an FGFR gene and a second previously independent gene or parts(s) of it. In some embodiments, the gene fusion may be an in-frame and/or an in-strand gene fusion with the partner gene. The breakpoint of the gene fusion (the point of fusion with the partner gene) may be located up- or downstream of the kinase domain. There are many possibilities as to why a fusion might activate the kinase, but one example is where the partner gene encodes a functional dimerization domain. Such fusion proteins may dimerize in the absence of an FGF ligand, thereby triggering activation of the FGFR kinase domain. Mutated FGFR2 genes which are FGFR2 gene fusions are of particular interest.

A gene truncation is another example of a gene rearrangement, e.g. an FGFR gene encoding a protein which has one or more amino acids missing from the N- or C-terminus. The truncation event may be upstream of exon 11 or in or downstream of exon 17, e.g. after amino acid 757 in FGFR1, after amino acid 761 in FGFR2 and after amino acid 751 in FGFR3.

Short variants include substitutions, insertions and deletions (indels). Base pair substitutions may result in a missense mutation (a change in a codon, resulting in an amino acid substitution in the protein), or a nonsense mutation (a premature stop codon, resulting in a truncated protein). Indels (small insertions or deletions) may result in a frameshift likely leading to a premature stop codon or may lead to incorporation of additional amino acids or loss of amino acids from the protein.

Some tumor FGFR genes may encode one or more amino acid substitutions, in particular in the Igll and/or Iglll region of the protein. In FGFR1 the Igll region is amino acids 158-246 and the Iglll region is amino acids 255-377 (Babina et al.). The corresponding regions in FGFR2 are 159-247 and 256-378. The corresponding regions in FGFR3 are 156-244 and 253-376.

Based on the Examples below, amino acid substitutions of interest include R254W in FGFR1 (with respect to SEQ ID NO: 2), Y244C in FGFR2 and G224S in FGFR3 (with respect to both SEQ ID NO: 4 and 10 for FGFR2 and with respect to both SEQ ID NO: 6 and 12 for FGFR3). Corresponding mutations in FGFR1, FGFR2 and FGFR3 are also of interest.

In some embodiments the protein encoded by a mutated FGFR1 gene may have one or more amino acid substitutions in the Igll region and/or Iglll region, namely in the region encompassed by amino acids 158- 246 and 255-377 (including the boundary amino acids, i.e. amino acids 158, 246, 255 and 377). The encoded FGFR1 protein may have an amino acid substitution at Y243 and/or G226. The encoded FGFR1 protein may have an amino acid substitution Y243C and/or G226S. (All amino acid positions are with respect to SEQ ID NO: 2.) In some embodiments the protein encoded by a mutated FGFR2 gene may have one or more amino acid substitutions in the Igll region and/or Iglll region, namely in the region encompassed by amino acids 159- 247 and 256-378 (including the boundary amino acids). The encoded FGFR2 protein may have an amino acid substitution at Y244 and/or G227. The encoded FGFR2 protein may have an amino acid substitution Y244C and/or G227S. (All amino acid positions are with respect to SEQ ID NO: 4.)

In some embodiments the protein encoded by a mutated FGFR3 gene may have one or more amino acid substitutions in the Igll region and/or Iglll region, namely in the region encompassed by amino acids 156- 244 and 253-376 (including the boundary amino acids). The encoded FGFR3 protein may have an amino acid substitution at Y241 and/or G224. The encoded FGFR3 protein may have an amino acid substitution Y241C and/or G224S. (All amino acid positions are with respect to SEQ ID NO: 6.)

In in vitro mouse models the following FGFR2 mutations with respect to SEQ ID NO: 4 have been found to be sensitive to derazantinib : C382R, Y375C, P253R, A97T , K569E, E767, E777del, E777K and F276C. The equivalent mutations in FGFR1 with respect to SEQ ID NO: 2 where there is the same amino acid at the equivalent position are C381R, Y374C, P252R, K566E, E774del, E774K and F275C. The equivalent mutations in FGFR3 with respect to SEQ ID NO: 6 where there is the same amino acid at the equivalent position are Y373C, P250R, A99T, K560E, E758, E768del, E768K and F273C.

In in vivo PDX/CDX models (CrownBio, Inc.) the following mutations have been found to be associated with sensitivity to derazantinib: I8S (FGFR2, model BL5001, bladder cancer); A97T (FGFR2, model 5384, bladder cancer); P286H and R285W (FGFR1, model GA1224, gastric cancer - equivalent to P255H and R254W in the FGFR1 amino acid sequence disclosed herein as SEQ ID NO: 2); GAB2 (FGFR2 fusion, model BRI 115, breast cancer); R671L (FGFR3, model HN3474, head and neck cancer - equivalent to R669L in the FGFR3 amino acid sequence disclosed herein as SEQ ID NO: 6); S783R and S789C (FGFR3, model LU5147, lung cancer - equivalent to S781R and S787C in the FGFR3 amino acid sequence disclosed herein as SEQ ID NO: 6); COL14A1 (FGFR2 fusion, model NCI-H716, colorectal cancer; model RT4, bladder cancer); TACC3 (FGFR3 fusion, model LU6426, lung cancer); and C382R (FGFR2, model UT5321, endometrial cancer).

The equivalent amino acid mutations in FGFR1 with respect to SEQ ID NO: 2 where there is the same amino acid at the equivalent position are P255H, R254W, R675L and S794C. The equivalent amino acid mutations in FGFR2 with respect to SEQ ID NO: 4 where there is the same amino acid at the equivalent position are I8S, P256H, R255W, R678L and S796C. The equivalent amino acid mutations in FGFR3 with respect to SEQ ID NO: 6 where there is the same amino acid at the equivalent position are P253H, R252W, R669L, S781R and S787C. Accordingly in some embodiments the encoded FGFR1 protein may harbor one or more amino acid mutations selected from C381R, Y374C, P252R, K566E, E774del, E774K, F275C, P255H, R254W, R675L and S794C, each with respect to SEQ ID NO: 2. In some embodiments the encoded FGFR2 protein may harbor one or more amino acid mutations selected from C382R, Y375C, P253R, A97T , K569E, E767, E777del, E777K, F276C, I8S, P256H, R255W, R6P, R678L and S796C, each with respect to SEQ ID NO: 4. In some embodiments the encoded FGFR3 protein may harbor one or more amino acid mutations selected from Y373C, P250R, A99T, K560E, E758, E768del, E768K, F273C, P253H, R252W, R669L, S781R and S787C each with respect to SEQ ID NO: 6. In some embodiments the encoded FGFR1, FGFR2 and/or FGFR3 encoded protein may be a fusion protein selected from TACC3, GAB2 and COL14Al.

In in vitro mouse models the following FGFR2 mutations with respect to SEQ ID NO: 4 have been found to be insensitive to derazantinib: V564F, K641R and N549K. The equivalent mutations in FGFR1 with respect to SEQ ID NO: 2 are V561F, K638R and N546K. The equivalent mutations in FGFR3 with respect to SEQ ID NO: 6 are V555F, K632R and N540K.

In models in vivo (CrownBio, Inc.) the following mutations have been found to be associated with insensitivity to derazantinib: G141V (FGFR1, model BL3578, bladder cancer - equivalent to G108V in the FGFR1 amino acid sequence disclosed herein as SEQ ID NO: 2); P803S (FGFR1, model BL5422, bladder cancer - equivalent to P772S in the FGFR1 amino acid sequence disclosed herein as SEQ ID NO: 2); G372C (FGFR3, model BL5425, bladder cancer - equivalent to G370C in the FGFR3 amino acid sequence disclosed herein as SEQ ID NO: 6); V218M (FGFR3, model GA6831, gastric cancer); P250L (FGFR3, model GA6889, gastric cancer); P504H (FGFR3, model GA6894, gastric cancer - equivalent to P502H in the FGFR3 amino acid sequence disclosed herein as SEQ ID NO: 6); N294I (FGFR3, model GA3155, gastric cancer); R3-POLN (FGFR3 fusion, model CR2054, colorectal cancer); S267P (FGFR2, model HN1420, head and neck cancer); JAKMIP1 (FGFR3 fusion, model LU5216, lung cancer); MCU (FGFR2 fusion, model LU0743, lung cancer); TACC3 (FGFR3 fusion, model RT112/84, bladder cancer); N549K (FGFR2, model AN3CA, MFE296, endometrial cancer); E636K (FGFR2, model A375, melanoma); and GJB2 (FGFR3 fusion, model BXF2211, bladder cancer).

The equivalent mutations in FGFR1 with respect to SEQ ID NO: 2 where there is the same amino acid at the equivalent position are G108V, P772S, V220M, P252L, N296I, E633K. The equivalent mutations in FGFR2 with respect to SEQ ID NO: 4 where there is the same amino acid at the equivalent position are P775S, V221M, P253L, N297I, S267P, E636K. The equivalent mutations in FGFR3 with respect to SEQ ID NO: 6 where there is the same amino acid at the equivalent position are P766S, G370C, V218M, P250L, P502H, N294I, S264P, E627K The amino acid positions in a given FGFR amino acid sequence corresponding to those indicated above can be determined by a person skilled in the art using the National Institute of Health Basica Local Alignment Search Tool (BLAST) using the default parameters, available at https://blast.ncbi.nlm.nih.gov/Blast.cgi.

A mutated FGFR gene may be expressed (in terms of mRNA) at a higher level than the corresponding matched-tissue non-tumor FGFR gene, in addition to carrying a mutation.

Within an FGFR gene, introns are enumerated based on the preceding exons according to coding sequence orientation, e.g.

• 5’ Exon 1- Intron 1- Exon 2 - Intron 2 - Exon 3 - Intron 3 - Exon 4 3’

• 3’ Exon 4 - Intron 3 - Exon 3 - Intron 2 - Exon 2 - Intron 1- Exon 1 5’

FGFR amplification and expression

A mutated FGFR gene may be amplified in terms of gene copy number at a higher level than the corresponding matched-tissue non-tumor FGFR gene. A standard value may be used for the mRNA expression level of the corresponding matched-tissue non-tumor FGFR gene. Amplification in normal cells is usually around two gene copies per cell (mean average) and may be taken as a standard value.

In some embodiments a mutated FGFR1 gene has an mRNA expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR1 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater. In some embodiments a mutated FGFR1 gene has a gene amplification level which is at least 6 gene copies per cell (mean average), e.g. at least 7 gene copies per cell (mean average), e.g. at least 8 gene copies per cell (mean average), e.g. at least 9 gene copies per cell (mean average), e.g. at least 10 gene copies per cell (mean average).

In some embodiments a mutated FGFR2 gene has an mRNA expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR2 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater. In some embodiments a mutated FGFR2 gene has a gene amplification level which is at least 6 gene copies per cell (mean average), e.g. at least 7 gene copies per cell (mean average), e.g. at least 8 gene copies per cell (mean average), e.g. at least 9 gene copies per cell (mean average), e.g. at least 10 gene copies per cell (mean average).

In some embodiments a mutated FGFR3 gene has an mRNA expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR3 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater. In some embodiments a mutated FGFR3 gene has a gene amplification level which is at least 6 gene copies per cell (mean average), e.g. at least 7 gene copies per cell (mean average), e.g. at least 8 gene copies per cell (mean average), e.g. at least 9 gene copies per cell (mean average), e.g. at least 10 gene copies per cell (mean average).

In the context of identifying patients with solid tumors (e.g. gastric adenocarcinoma) for treatment with derazantinib and paclitaxel on the basis of the solid tumor having a level of TAMs that is higher than a standard value, a mutated or wild type FGFR gene may be amplified in terms of gene copy number at a higher level than the corresponding matched-tissue non-tumor FGFR gene, e.g. higher than two gene copies per cell (mean average). Likewise, mRNA expression level of a mutated or wild type FGFR may be higher than the corresponding matched-tissue non-tumor FGFR gene.

Accordingly, in some embodiments a mutated or wild type FGFR1 gene has an mRNA expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR1 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater.

In some embodiments a mutated or wild type FGFR1 gene has an RNA-Seq(log2) mRNA expression level of at least 5, e.g. at least 5.5, e.g. at least 6, e.g. at least 6.5, e.g. at least 7. See e.g. Koch et al. Am J Respir Cell Mol Biol 2018, 59:2,145-157. In some embodiments a mutated or wild type FGFR1 gene has a gene amplification level which is at least 4 gene copies per cell (mean average), e.g. at least 6 gene copies per cell (mean average), e.g. at least 7 gene copies per cell (mean average), e.g. at least 8 gene copies per cell (mean average), e.g. at least 9 gene copies per cell (mean average), e.g. at least 10 gene copies per cell (mean average).

In some embodiments a mutated or wild type FGFR2 gene has an mRNA expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR2 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater. In some embodiments a mutated or wild type FGFR2 gene has an RNA-Seq(log2) mRNA expression level of at least 5, e.g. at least 5.5, e.g. at least 6, e.g. at least 6.5, e.g. at least 7. In some embodiments a mutated or wild type FGFR2 gene has a gene amplification level which is at least 4 gene copies per cell (mean average), e.g. at least 6 gene copies per cell (mean average), e.g. at least 7 gene copies per cell (mean average), e.g. at least 8 gene copies per cell (mean average), e.g. at least 9 gene copies per cell (mean average), e.g. at least 10 gene copies per cell (mean average). In some embodiments a mutated or wild type FGFR3 gene has an mRNA expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR3 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater. In some embodiments a mutated or wild type FGFR3 gene has an RNA-Seq(log2) mRNA expression level of at least 5, e.g. at least 5.5, e.g. at least 6, e.g. at least 6.5, e.g. at least 7. In some embodiments a mutated or wild type FGFR3 gene has a gene amplification level which is at least 4 gene copies per cell (mean average), e.g. at least 6 gene copies per cell (mean average), e.g. at least 7 gene copies per cell (mean average), e.g. at least 8 gene copies per cell (mean average), e.g. at least 9 gene copies per cell (mean average), e.g. at least 10 gene copies per cell (mean average).

In some embodiments the median average number of the gene copies per cell may be used in place of the mean average.

It is expected that higher levels of FGFR mRNA expression lead to higher levels of FGFR protein expression. In the context of identifying patients with solid tumors (e.g. gastric adenocarcinoma) for treatment with derazantinib and paclitaxel on the basis of the solid tumor having a level of TAMs that is higher than a standard value the protein expression level of a mutated or wild type FGFR may be higher than the corresponding matched-tissue non-tumor FGFR gene.

Accordingly, in some embodiments a mutated or wild type FGFR1 gene has a protein expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR1 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater. In some embodiments a mutated or wild type FGFR2 gene has a protein expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR2 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater. In some embodiments a mutated or wild type FGFR3 gene has a protein expression level which is at least about 50% greater than the matched-tissue non-tumor FGFR3 gene, e.g. at least about 100% greater, e.g. at least about 200% greater, e.g. at least about 300% greater, e.g. at least about 400% greater.

Gene expression and/or amplification is measured in a sample taken from the patient according to techniques known to the person skilled in the art, including those described below.

M2 Tumor Associated Macrophages

Macrophages are immunocytes, which perform a broad spectrum of functions that range from modulating tissue homeostasis, defending against pathogens, and facilitating wound healing. Macrophages infiltrating tumor tissues or populated in the microenvironment of solid tumors are known as tumor-associated macrophages and are generally associated with a worse disease outcome. Tumor-associated macrophages are known to have a role in tumor growth, tumor angiogenesis, immune regulation, metastasis, and chemoresistance.

Macrophages are capable of adopting different phenotypes, depending on the particular microenvironment. Activated macrophages are classified into the Ml (classical-activated macrophages) and M2 (alternative-activated macrophages) phenotype. Ml macrophages promote an inflammation response against invading pathogens and tumor cells, whereas M2 macrophages adopt an immune suppressive phenotype, favoring tissue repair and tumor progression. The two different types of macrophages can be distinguished based on their different markers, metabolic characteristics and gene expression profiles. Ml macrophages secrete proinflammatory cytokines such as IL-12, tumor necrosis factor (TNF)-a, CXCL-10, and interferon (IFN)-y and produce high levels of nitric oxide synthase, while M2 macrophages secrete anti-inflammatory cytokines such as IL-10, IL-13, and IL-4 and express abundant arginase-1, mannose receptor (MR, CD206), and scavenger receptors. The conversion between Ml (anti-tumorigenesis) and M2 (pro-tumorigenesis) is a biological process named “macrophage polarization” in response to microenvironmental signals. Though studies found that TAMs are able to exhibit either polarization phenotype, researchers tend to consider tumor-associated macrophages as M2- like phenotype-acquired macrophages (Lin et al. Journal of Hematology & Oncology 2019;12:76). Accordingly, reference herein to “Tumor associated macrophages” refers to tumor-associated macrophages displaying the M2 phenotype (TAMs) unless otherwise specified.

Biomarkers which may be used in the detection of TAMs include the following: MMP2/9, B7-H4, STAT- 3 HLA-DR, CD68, CD14, CD163, CD206, CD204, IL-10. In particular, MMP2/9, B7-H4, STAT-3 and CD163 are present on M2 macrophages but show no expression on Ml macrophages (Lin et al. Journal of Hematology & Oncology 2019;12:76). Arginase-1 may also be used as a biomarker for TAMs (Arlauckas et al. Theranostics. 2018; 8(21): 5842-5854). In some embodiments CD68 is used as a biomarker for detecting and/or determining the level of TAMs. In some embodiments CD 163 is used as a biomarker for detecting and/or determining the level of TAMs. In some embodiments Arginase- 1 is used as a biomarker for detecting and/or determining the level of TAMs.

TAMs may be detected and/or the level of TAMs may be determined/measured in tumor samples using well-known techniques, for example by gene expression analysis, immunohistochemistry, ELISA, flow cytometry etc. For more details of specific examples see Sawa-Wejksza K et al. Arch Immunol Ther Exp (Warsz) 2018;66(2):97-l 11; Martin et al. Cancer Metastasis Rev. 2007;26(3-4):717-24; Kryczek et al. J Exp Med. 2006;203(4):871-81; Yu et al. Nature Rev Cancer. 2009;9(l l):798-809; Heusinkveld et al. J Translat Med. 2011 ;9:216; Martinez et al. J Immunol. 2006;177(10):7303-l 1; and Verreck et al.. J LeukocBiol. 2006;79(2):285-93. TAMs may also be detected and/or determined/measured using non- invasive techniques, e.g. as described in Aghighi et al. Clin. Cancer Res. 2018;24: 17, which describes using ferumoxytol -enhanced magnetic resonance imaging of TAMs to identify patients with T AM-rich tumors.

TAMs can constitute up to 50% of the inflammatory cells in a solid tumor (Zhang et al. Pharmacological Research 161 (2020) 105111), whereas the median stroma component across a broad range of tumor types can be calculated as around 30% (Wu et al. Oncotarget, 2016, Vol. 7, No. 429). Based on this information from patient tumors, and taking into account the presence of non-inflammatory cells in the tumor, a maximum median TAM level could be estimated as 10% of the total number of cells in a tumor. In the animal models described in the Examples a mean/median average of 6-8% TAM cells are likely to be at the high-end, and a mean/median average of <1% TAM cells is at the low end. The level of TAM calls in a tumor as a percentage of the total number of cells in a tumor may be determined using electronic scanning immunohistochemical techniques, as described in the Examples.

As described herein, a higher level of TAMs relative to a standard value may be used to stratify patients for treatment with derazantinib and paclitaxel. As used herein, an increase or relatively high or high or higher levels relative to a standard value means the amount or concentration of the biomarker in a sample is detectably more in the sample relative to the standard value.

In some embodiments a tumor may be considered to have a higher level of TAMs relative to a standard value when the level of TAMs is determined to be a mean average of at least 1% of the total number of cells in a tumor sample. In some embodiments a tumor may be considered to have a higher level of TAMs relative to a standard value when the level of TAMs is determined to be a mean average of e.g. at least 2%, e.g. at least 3%, e.g. at least 4%, e.g. at least 5%, e.g. at least 6% of the total number of cells in a tumor sample. In some embodiments the median average level of TAMs be used in place of the mean average.

In some embodiments a tumor may be considered not to have a higher level of TAMs relative to a standard value when the level of TAMs is determined to be a mean average of less than 1% of the total number of cells in a tumor sample. In some embodiments a tumor may be considered not to have a higher level of TAM relative to a standard value when the level of TAMs is determined to be a mean average of e.g. less than 2%, e.g. less than 3%, e.g. less than 4%, e.g. less than 5%, e.g. less than 6% of the total number of cells in a tumor sample. In some embodiments the median average level of TAMs be used in place of the mean average.

In some embodiments a tumor may be considered to have a higher level of TAMs when it has at least an increase of, or higher level of, about 1% relative to the standard, e.g. at least an increase of about 5% relative to the standard. For example it is an increase of, or higher level of, at least about 10% relative to the standard. For example it is an increase of, or higher level of, at least about 20% relative to the standard. For example, such an increase of, or higher level of, may include, but is not limited to, at least about 1%, at least about 10%, at least about 20%, at least about 30%, at least about 50%, at least about 70%, at least about 80%, at least about 90% or at least about 100% or > 100% increase relative to the standard.

Fujiwara et al. (Fujiwara et al. Am J Pathol. 2011 Sep; 179(3): 1157-1170) describes an immunohistochemical method for measuring/determining the level of TAMs in a tumor sample, which may be employed: An antibody suitable for detecting a TAM biomarker (e.g. CD68) may be used to evaluate a clinical sample. Samples may be fixed in 10% neutral buffered formalin and embedded in paraffin. After the sections are deparaffinized in xylene and rehydrated in a graded ethanol series, they may be subjected to microwave pretreatment with citrate buffer (pH 6.0). After incubation with the antigen-specific antibody, samples may be incubated with horseradish peroxidase-labeled goat antimouse antibodies. The reaction may be visualized using the diaminobenzidine substrate system and the samples may be counterstained with diluted hematoxylin. To count the macrophages, an image with an area of 0.64 mm² may be created from six different visual fields. The level of TAMs in a tumor sample may be considered at the mean average number of cells positive for the biomarker versus the total number of cells per visual field counted over six random fields.

In some embodiments the methods of the invention include the step of determining the level of TAMs ex vivo in a sample taken from the patient, preferably a tissue sample taken from the solid tumor.

In some embodiments determining that the level of TAMs in the solid tumor is higher than a standard value involves determining the level of TAMs in a sample taken from the patient i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

If the level of TAMs in the tumor sample is higher than the level defined in i), ii) or iii) then the level of TAMs in the tumor sample is considered to be higher than the standard value.

In one preferred embodiment the standard value (or a set of standard values) are established from samples from a population of subjects with the same tumor histotype. The standard value may be established ex- vivo from pre-obtained samples which may be from cell lines or animal tumor models, or preferably biological material from at least one human subject and more preferably from an average (e.g. mean average) of subjects (e.g., n=2 to 1000 or more). The standard value may then be correlated with the response data of the same cell lines, or same subjects, to treatment with derazantinib and paclitaxel. From this correlation a comparator module, for example in the form of a relative scale or scoring system, optionally including cut-off or threshold values, can be established which indicates the levels of TAMs / gene amplification / gene expression associated with a spectrum of response to the combination of derazantinib and paclitaxel. In a preferred embodiment this comparator module comprises a cut-off value or set of values which predicts synergistic response of the combination to treatment using the Clarke- Index (see Examples below).

For example, if an immunohistochemical method is used to measure the level of TAMs in a sample, standard values may be in the form of a scoring system. Such a system might take into account the percentage of cells in which staining for TAM is present. The system may also take into account the relative intensity of staining in the individual cells. The standard values or set of standard values of the level of TAMs may then be correlated with data indicating the response, especially whether or not the response is synergistic, of the subject or tissue or cell line to the therapeutic activity of the combination of derazantinib and paclitaxel.

In a further aspect the invention provides a method of selecting a patient with a solid tumor (e.g. gastric adenocarcinoma) for treatment with derazantinib and paclitaxel, wherein the solid tumor has at least one of the following characteristics:

(iii) the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value; the method comprising the steps of:

(a) determining ex vivo the level of TAMs in a sample of the solid tumor taken from the patient; and

(b) selecting a patient for treatment with derazantinib and paclitaxel if the level of TAMs in the sample is higher than a standard value.

In some embodiments the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene. In some embodiments the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value. In some embodiments the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

Optionally the method includes one of the following steps:

(cl) treating the patient with a therapeutically effective amount of derazantinib and paclitaxel; (c2) treating the patient with a therapeutically effective amount of derazantinib, which patient is undergoing or will undergo treatment with paclitaxel; and

(c3) treating the patient with a therapeutically effective amount of paclitaxel, which patient is undergoing or will undergo treatment with derazantinib .

In some embodiments the method includes prior to step (b): determining ex vivo in a sample taken from the patient whether the patient’s solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

In some embodiments the method includes prior to step (b): determining ex vivo in a sample taken from the patient whether the patient’s solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

In some embodiments the method includes prior to step (b): determining ex vivo in a sample taken from the patient whether the patient’s solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

Tumors

The solid tumors to be treated by the present invention will usually be malignant tumors, e.g. cancers. Examples of cancers which may be treated using the methods of the present invention include cancers in terms of the organs and parts of the body affected include, but are not limited to, the breast, cervix, ovaries, colon, rectum (including colon and rectum i.e. colorectal cancer), lung (including small cell lung cancer, non-small cell lung cancer, large cell lung cancer and mesothelioma), endocrine system, bone, adrenal gland, thymus, liver, stomach, intestine, pancreas, bone marrow, bladder, urinary tract, kidneys, skin, thyroid, brain, head, neck, prostate and testis. In some embodiments the cancer is selected from the group consisting of breast cancer, prostate cancer, cervical cancer, ovarian cancer, gastric adenocarcinoma, colorectal cancer, pancreatic cancer, liver cancer, brain cancer, neuroendocrine cancer, lung cancer, kidney cancer, bladder cancer, mesothelioma, melanomas and sarcomas.

In some embodiments the solid tumor is gastric adenocarcinoma, or is a urothelial cancer or intrahepatic cholangiocarcinoma. Preferably the solid tumor is gastro adenocarcinoma.

Samples

The FGFR mutational status of a patient’s tumor may be determined ex vivo (in vitro) on a sample of biological material derived from the patient. The sample may be any biological material separated from the body such as, for example, a tissue sample and/or a fluid sample. In some embodiments, the sample is taken from tumor tissue, e.g. a tumor biopsy. In some embodiments, the sample is a fluid sample such as a blood sample, which may be processed e.g. to provide a plasma or serum sample, or a urine, saliva, cerebrospinal fluid, pleural effusion or bile sample.

Methods for removal of samples are well known in the art, e.g. by surgical resection or by biopsy, for example by punch biopsy, core biopsy or aspiration fine needle biopsy, endoscopic biopsy, surface biopsy or liquid biopsy. Liquid biopsy is a term usually used in contraposition to the traditional “solid” tissue biopsy and is based on the finding that tumor-derived material is shedded into different bio-fluids from where it can be isolated and analyzed (Arechederra et al. Adv Lab Med 2020;20200009). When performing liquid biopsy the tumor DNA and /or RNA may be extracted from the liquid sample, e.g. from circulating tumor cells, circulating tumor DNA, circulating tumor RNA and/or exosomes.

In some embodiments, the tumor DNA is extracted from a patient’s blood sample and sequenced to determine the FGFR mutational status of the patient’s tumor. In some embodiments, the tumor mRNA is extracted from a patient’s blood sample and sequenced to determine the FGFR mutational status of the patient’s tumor. In some embodiments, the tumor DNA is extracted from a patient’s tissue sample and sequenced to determine the FGFR mutational status of the patient’s tumor. In some embodiments, the tumor mRNA is extracted from a patient’s tissue sample and sequenced to determine the FGFR mutational status of the patient’s tumor. In some embodiments, the patient’s normal FGFR DNA sequence may be determined by taking DNA from non-tumor cells, e.g. from matched tissue non-tumor cells.

The presence of a mutated FGFR gene may be identified using standard techniques and may be detected at the nucleic and and/or the protein level. Typical techniques of detecting mutated genes at the nucleic acid level include but are not limited to: i) using a labelled probe that is capable of hybridizing to mRNA or DNA (e.g. fluorescence in situ hybridization (FISH), which is a well-known tool for detecting specific biomarkers in neoplasms, see for example Hu L. Biomarker Research 2014;2: 1, 3); ii) using polymerase chain reaction (PCR) involving one or more primers based on the FGFR gene sequence, for example using real-time PCR (RT-PCR), quantitative PCR methods using labelled probes, e.g. Anorogenic probes, such as quantitative real-time PCR, as well as next generation sequencing techniques; iii) micro-arrays; iv) northern blotting. As is well understood in the art, similar techniques may be used to determine FGFR gene amplification and/or expression (e.g. mRNA expression), in particular PCR or micro array based techniques, e.g. the nCounter® approach developed by NanoString Technologies, Inc..

Typical techniques of detecting genes (including mutated genes) at the protein level include but are not limited to i) immunohistochemistry (IHC) analysis; ii) western blotting; iii) immunoprecipitation; iv) enzyme linked immunosorbent assay (ELISA); v) radioimmunoassay; vi) fluorescence activated cell sorting (FACS); vii) mass spectrometry, including matrix assisted laser desorption/ionization (MALDI, e.g. MALDI-TOF) and surface enhanced laser desorption/ionization (SELDI, e.g. SELDI-TOF).

Some of the above methods require antibodies capable of selectively binding to mutant proteins and/or wild type proteins. The antibodies involved may be monoclonal or polyclonal antibodies, antibody fragments, and/or various types of synthetic antibodies, including chimeric antibodies, DARPS (designed Ankyrin repeat proteins) or DNA/RNA aptamers. The antibody may be labelled to enable it to be detected or capable of detection following reaction with one or more further species, for example using a secondary antibody that is labelled or capable of producing a detectable result. Antibodies can be prepared via conventional antibody generation methods well known to a skilled person.

Patient samples to determine the level of TAMs in the solid tumor are preferably taken as tissue samples from the tumor.

Administration

Administration of the pharmaceutical combinations of the invention includes administration of the combination in a single formulation or unit dosage form, as well as administration of the individual agents of the combination in separate formulations or separate dosage forms.

In some embodiments, the combination of the invention is used for the treatment of gastric adenocarcinoma comprising administering to the subject a combination therapy, comprising a therapeutically effective amount of derazantinib and a therapeutically effective amount of paclitaxel. These compounds are administered at therapeutically effective dosages, which when combined may provide a beneficial effect e.g. as described herein. The skilled person will understand that therapeutically effective dosages for use in combination therapy may be lower than the dosages required to provide a therapeutic effect when using either agent as a monotherapy.

The administration of a pharmaceutical combination of the invention may result not only in a beneficial effect, e.g. a synergistic effect, e.g. with regard to alleviating, delaying progression of or inhibiting the symptoms, but may also result in further beneficial effects, e.g. fewer side-effects, more durable therapeutic effect, an improved quality of life or a decreased morbidity, compared with a monotherapy applying only one of the pharmaceutically therapeutic agents used in the combination of the invention. It may also be the case that lower doses of the therapeutic agents of the combination of the invention can be used, for example, such that the dosages may not only often be smaller, but also may be applied less frequently, or can be used in order to diminish the incidence of side-effects observed with one of the combination partners alone. In some embodiments, the combination provided herein may display a synergistic effect. The term "synergistic effect" as used herein, refers to action of the two agents derazantinib and paclitaxel, to produce a therapeutic effect, e.g. slowing the progression of the disease or symptoms thereof, which is greater than the addition of the same therapeutic effect of each drug administered on its own.

Generally, in determining a synergistic interaction between one or more components, the optimum range for the effect and absolute dose ranges of each component for the effect may be definitively measured by administration of the components over different w/w ratio ranges and doses to patients in need of treatment. For humans, the complexity and cost of carrying out clinical studies on patients may render impractical the use of this form of testing as a primary model for synergy. However, the observation of synergy in certain experiments can be predictive of the effect in other species, and animal models may be used to further quantify a synergistic effect. The results of such studies can also be used to predict effective dose ratio ranges and the absolute doses and plasma concentrations, e.g. as illustrated in the Examples below.

In further embodiments, the present invention provides a synergistic combination for administration to humans comprising the pharmaceutical combination of the invention, where the dose range of each component corresponds to the synergistic ranges, e.g. as indicated in a suitable tumor model or clinical study.

The combinations of the present invention can be used in long-term therapy or as an adjuvant therapy in the context of other treatment strategies, as described above. Other possible treatments are therapy to maintain the patient's status after tumor regression, or even preventive therapy, for example in patients at risk.

Derazantinib and paclitaxel may be administered according to the same treatment schedule or may be administered according to independent treatment schedules. The treatment schedules may be cyclic or continuous.

A cyclic treatment schedule is defined by a repeated dosing schedule wherein the repeated element (a cycle) has a specific duration and wherein doses are administered on specific days within the cycle. A cycle may incorporate a period, usually at the end of the cycle, in which there is no administration (a “rest period”), e.g. to allow a period for recovery. A treatment cycle may be, e.g. 7 days, 14 days, 21 days, 28 days or longer. A continuous treatment schedule is a regular dosing schedule, which does not incorporate rest periods (i.e. periods that are longer than the regular interval between the doses). For example, doses may be administered once per day, twice per day, once every two days, once every three days etc. The treatment schedule, whether cyclic or continuous may be continued for as long as required (an “open-end treatment”) e.g. as long as the patient is receiving benefit judged by a physician overseeing the treatment.

When derazantinib and paclitaxel are administered according to independent treatment schedules, the treatment schedules may both be cyclic, or one may be cyclic and the other may be continuous. When both treatment schedules are cyclic, the cycles of the two treatment schedules may be of the same duration or may be of different duration, and they may start on the same day or may start on different days.

In some embodiments, derazantinib is administered according to a continuous treatment schedule. In some embodiments, derazantinib is administered according to a continuous treatment schedule in which derazantinib is administered once per day. In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered once per day. In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered twice per day.

In some embodiments, paclitaxel is administered according to a cyclic treatment schedule. In some embodiments, paclitaxel is administered according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15.

In some embodiments, derazantinib is administered according to a continuous treatment schedule and paclitaxel is administered according to a cyclic treatment schedule. In some embodiments, derazantinib is administered according to a continuous treatment schedule in which derazantinib is administered once per day and paclitaxel is administered according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15. In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered once per day and paclitaxel is administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15.

In some embodiments, derazantinib is administered according to a continuous treatment schedule in which derazantinib is administered twice per day and paclitaxel is administered according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15. In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered twice per day and paclitaxel is administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15.

Effective dosages of each of the combination partners employed in the combinations of the invention may vary depending on the pharmaceutical composition employed, the mode of administration, the condition being treated, and the severity of the condition being treated. Thus, the dosage regimen of the combination of the invention is selected in accordance with a variety of factors including the route of administration and the renal and hepatic function of the patient.

The optimum ratios, individual and combined dosages, and concentrations of the combination partners of the pharmaceutical combination of the invention that yield efficacy without toxicity are based on the kinetics of the therapeutic agents' availability to target sites. They may be established using routine clinical testing and procedures that are well known in the art and will depend upon a variety of factors, such as the mode of administration, the condition being treated and the severity of the condition being treated, as well as the age, body weight, general health, gender and diet of the individual and other medications the individual is taking. Likewise, frequency of dosage may vary depending on the compound used and the particular condition to be treated. Patients may generally be monitored for therapeutic effectiveness using assays suitable for the condition being treated, which will be familiar to those of ordinary skill in the art.

Example daily oral dosage amounts of derazantinib include about 50 mg to about 400 mg, e.g. about 50 mg to about 300 mg, e.g. about 100 mg to about 300 mg, e.g. about 150 mg to about 300 mg, e.g. about 200 mg to about 300 mg. Further example daily oral dosage amounts of derazantinib include about 50 mg to about 300 mg, e.g. about 100 mg to about 400 mg, e.g. about 150 mg to about 400 mg, e.g. about 200 mg to about 400 mg, e.g. about 300 mg to about 400 mg.

Examples of specific daily oral dosages amounts include about 50 mg, about 60 mg, about 70 mg, about 80 mg, about 90 mg, about 100 mg, about 110 mg, about 120 mg, about 130 mg, about 140 mg, about 150 mg, about 160 mg, about 170 mg, about 180 mg, about 190 mg, about 200 mg, about 210 mg, about 220 mg, about 230 mg, about 240 mg, about 250 mg, about 260 mg, about 270 mg, about 280 mg, about 290 mg, about 300 mg, about 310 mg, about 320 mg, about 330 mg, about 340 mg, about 350 mg, about 360 mg, about 370 mg, about 380 mg, about 390 mg, about 400 mg,

Example weekly intravenous dosage amounts of paclitaxel include about 60 mg/m²to about 90 mg/m², e.g. about 60 mg/m² to about 80 mg/m², e.g. about 70 mg/m² to about 80 mg/m². Examples of specific weekly intravenous dosage amounts include about 60 mg/m², about 61 mg/m², about 62 mg/m², about 63 mg/m², about 64 mg/m², about 65 mg/m², about 66 mg/m², about 67 mg/m², about 68 mg/m², about 69 mg/m², about 70 mg/m², about 71 mg/m², about 72 mg/m², about 73 mg/m², about 74 mg/m², about 75 mg/m², about 76 mg/m², about 77 mg/m², about 78 mg/m², about 79 mg/m², about 80 mg/m², about 81 mg/m², about 82 mg/m², about 83 mg/m², about 84 mg/m², about 85 mg/m², about 86 mg/m², about 87 mg/m², about 88 mg/m², about 89 mg/m², about 90 mg/m².

In some embodiments, the daily oral dosage amounts of derazantinib is about 50 mg to about 400 mg and the weekly intravenous dosage amounts of paclitaxel is about 60 mg/m² to about 90 mg/m². In some embodiments, the daily oral dosage amounts of derazantinib is about 50 mg to about 300 mg and the weekly intravenous dosage amounts of paclitaxel is about 60 mg/m² to about 90 mg/m². In some embodiments, the daily oral dosage amounts of derazantinib is about 100 mg to about 300 mg and the weekly intravenous dosage amounts of paclitaxel is about 60 mg/m² to about 80 mg/m². In some embodiments, the daily oral dosage amounts of derazantinib is about 200 mg to about 300 mg and the weekly intravenous dosage amounts of paclitaxel is about 70 mg/m² to about 80 mg/m².

In some embodiments, the daily oral dosage amounts of derazantinib is about 100 mg to about 400 mg and the weekly intravenous dosage amounts of paclitaxel is about 60 mg/m² to about 80 mg/m². In some embodiments, the daily oral dosage amounts of derazantinib is about 200 mg to about 400 mg and the weekly intravenous dosage amounts of paclitaxel is about 70 mg/m² to about 80 mg/m². In some embodiments, the daily oral dosage amounts of derazantinib is about 300 mg to about 400 mg and the weekly intravenous dosage amounts of paclitaxel is about 70 mg/m² to about 80 mg/m².

In some embodiments, the daily oral dosage amounts of derazantinib is about 50 mg to about 400 mg and the weekly intravenous dosage amounts of paclitaxel is about 60 mg/m² to about 90 mg/m², in which paclitaxek is administered according to a 28-day cyclic treatment schedule, dosing on days 1, 8 and 15. In some embodiments, the daily oral dosage amounts of derazantinib is about 50 mg to about 300 mg and the weekly intravenous dosage amounts of paclitaxel is about 60 mg/m² to about 90 mg/m², in which paclitaxek is administered according to a 28-day cyclic treatment schedule, dosing on days 1, 8 and 15. In some embodiments, the daily oral dosage amounts of derazantinib is about 100 mg to about 300 mg and the weekly intravenous dosage amounts of paclitaxel is about 60 mg/m² to about 80 mg/m², in which paclitaxek is administered according to a 28-day cyclic treatment schedule, dosing on days 1, 8 and 15. In some embodiments, the daily oral dosage amounts of derazantinib is about 200 mg to about 300 mg and the weekly intravenous dosage amounts of paclitaxel is about 70 mg/m² to about 80 mg/m² , in which paclitaxek is administered according to a 28-day cyclic treatment schedule, dosing on days 1, 8 and 15.

In some embodiments, the daily oral dosage amounts of derazantinib is about 100 mg to about 400 mg and the weekly intravenous dosage amounts of paclitaxel is about 60 mg/m² to about 80 mg/m², in which paclitaxek is administered according to a 28-day cyclic treatment schedule, dosing on days 1, 8 and 15. In some embodiments, the daily oral dosage amounts of derazantinib is about 200 mg to about 400 mg and the weekly intravenous dosage amounts of paclitaxel is about 70 mg/m² to about 80 mg/m² , in which paclitaxek is administered according to a 28-day cyclic treatment schedule, dosing on days 1, 8 and 15. In some embodiments, the daily oral dosage amounts of derazantinib is about 300 mg to about 400 mg and the weekly intravenous dosage amounts of paclitaxel is about 70 mg/m² to about 80 mg/m² , in which paclitaxek is administered according to a 28-day cyclic treatment schedule, dosing on days 1, 8 and 15.

Additional embodiments are presented in Table 1 below. Table 1

*In each embodiment 1 to 1116, derazantinib is administered orally at the indicated dose (continuous treatment schedule dosing every day) and paclitaxel is administered intravenously at the indicated dose (28-day cyclic treatment schedule, dosing on days 1, 8 and 15).

In some embodiments, the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be about 0.3: 1.0 to about 3.6: 1.0, e.g. about 0.6: 1.0 to about 3.1: 1.0, e.g. about 1.1 : 1.0 to about 3. 1 : 1.0, e.g. about 1.3 : 1.0 to about 2.7: 1.0. In some embodiments, the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be at least about 0.3: 1.0, e.g. at least about 0.6: 1.0, e.g. at least about 1.1: 1.0, e.g. at least about 1.3: 1.0. In some embodiments, the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be up to about 3.6: 1.0, e.g. up to about 3. 1: 1.0, e.g. up to about 2.7: 1.0.

In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered once per day and paclitaxel is administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15 and wherein the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be about 0.3: 1.0 to about 3.6: 1.0, e.g. about 0.6: 1.0 to about 3. 1: 1.0, e.g. about 1.1: 1.0 to about 3.1: 1.0, e.g. about 1.3: 1.0 to about 2.7: 1.0.

In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered once per day and paclitaxel is administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15 and wherein the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be at least about 0.3: 1.0, e.g. at least about 0.6: 1.0, e.g. at least about 1.1: 1.0, e.g. at least about 1.3: 1.0. In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered once per day and paclitaxel is administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15 and wherein the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be up to about 3.6: 1.0, e.g. up to about 3. 1: 1.0, e.g. up to about 2.7: 1.0.

In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered twice per day and paclitaxel is administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15 and wherein the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be about 0.3: 1.0 to about 4.8: 1:0, e.g. about 0.6: 1.0 to about 4.1: 1, e.g. about 1.1: 1.0 to about 4.1: 1.0, e.g. about 1.3: 1.0 to about 3.6: 1.0.

In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered twice per day and paclitaxel is administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15 and wherein the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be at least about 0.3: 1.0, e.g. at least about 0.6: 1.0, e.g. at least about 1.1: 1.0, e.g. at least about 1.3: 1.0.

In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered twice per day and paclitaxel is administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15 and wherein the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg may be up to about 4.8: 1:0, e.g. up to about 4.1: 1, e.g. up to about 4.1: 1.0, e.g. up to about 3.6: 1.0.

The method of treating gastric adenocarcinoma according to the invention may comprise (i) administration of derazantinib in free or pharmaceutically acceptable salt form and (ii) administration of paclitaxel in free or pharmaceutically acceptable salt form simultaneously or sequentially in any order, in jointly therapeutically effective amounts, e.g. in synergistically effective amounts, e.g. in continuous or cyclic dosing schedule corresponding to the amounts described herein. The individual combination partners of the combination of the invention may be administered separately at different times during the course of therapy or concurrently. The invention is therefore to be understood as embracing all such regimens of simultaneous or alternating treatment and the term "administering" is to be interpreted accordingly. When administered on the same day derazantinib will usually be administered first.

Generally, derazantinib and paclitaxel will be administered at dosages that do not exceed the maximum tolerated dose (MTD) for a particular mode of administration and indication, as determined in a clinical dose escalation study.

Synthesis

Derazantinib may be synthesized starting from (4R)-4-(2-fluorophenyl)tetralin-l-one according to the procedures described in WO 2017/106639. (4R)-4-(2-fluorophenyl)tetralin-l-one may be synthesized from (2-fluorophenyl)-phenyl-methanone via (E/Z)-3-ethoxycarbonyl-4-(2-fluorophenyl)-4-phenyl-but-3- enoic acid and asymmetric hydrogenation of (E/Z)-4-(2-fluorophenyl)-4-phenyl-but-3-enoic acid using a rhodium-phanephos catalyst analogously to the methodology described in Boulton et al., Org. Biomol. Chem. 2003;1: 1094-1096, followed by cyclisation. The intermediate l-[3-(2- hydroxyethyl)phenyl]guanidine may be prepared by reduction of 2-(3-nitrophenyl)acetic acid e.g. with sodium borohydrate / boron trifluoride etherate to give 2-(3-nitrophenyl)ethan-l-ol followed by hydrogenation e.g. with raney nickel to give 2-(3-aminophenyl)ethanol followed by reaction with cyanamide, optionally in the presence of methanesulfonic acid to give the product as the methane sulfonic acid salt.

Formulations

The compounds of the invention may be formulated for administration according to procedures known to the person skilled in the art. For example compositions may be formulated for non-parenteral administration, such as nasal, buccal, rectal, pulmonary, vaginal, sublingual, topical, transdermal, ophthalmic, or, especially, for oral administration, e.g. in the form of oral solid dosage forms, e.g. granules, pellets, powders, tablets, film or sugar-coated tablets, effervescent tablets, hard and soft gelatin or hydroxypropylmethylcellulose (HPMC) capsules, coated as applicable, orally disintegrating tablets, oral solutions, lipid emulsions or suspensions, or for parenteral administration, such as intravenous, intramuscular, or subcutaneous, intrathecal, intradermal or epidural administration, to mammals, especially humans, e.g. in the form of solutions, lipid emulsions or suspensions containing microparticles or nanoparticles.

The compounds of the invention can be processed with pharmaceutically inert, inorganic or organic excipients for the production of oral solid dosage forms, e.g. granules, pellets, powders, tablets, film or sugar coated tablets, effervescent tablets, hard gelatin or HPMC capsules or orally disintegrating tablets. Fillers e.g. lactose, cellulose, mannitol, sorbitol, calcium phosphate, starch or derivatives thereof, binders e.g. cellulose, starch, polyvinylpyrrolidone, or derivatives thereof, glidants e.g. talcum, stearic acid or its salts, flowing agents e.g. fumed silica, can be used as such excipients for formulating and manufacturing of oral solid dosage forms, such as granules, pellets, powders, tablets, fdm or sugar coated tablets, effervescent tablets, hard gelatin or HPMC capsules, or orally disintegrating tablets. Suitable excipients for soft gelatin capsules are e.g. vegetable oils, waxes, fats, semisolid and liquid polyols etc.

Suitable excipients for the manufacture of oral solutions, lipid emulsions or suspensions are e.g. water, alcohols, polyols, saccharose, invert sugar, glucose etc. Suitable excipients for parenteral formulations are e.g. water, alcohols, polyols, glycerol, vegetable oils, lecithin, surfactants etc. Moreover, the pharmaceutical preparations can contain preservatives, solubilizers, stabilizers, wetting agents, emulsifiers, sweeteners, colorants, flavorants, salts for varying the osmotic pressure, buffers, masking agents or antioxidants. They can also contain other therapeutically valuable substances.

The paclitaxel formulation sold under the brand name Taxol® contains Cremophor® EL (polyethoxylated castor oil) and ethanol which is diluted with a suitable parenteral fluid for intravenous administration. Alternative paclitaxel formulations include those such as:

• Abraxane®: a nanoparticle formulation of paclitaxel and human serum albumin

• Taxoprexin®: a prodrug of paclitaxel chemically bound to the fatty acid, docosahexaenoic acid

• Paclical® poliglumex: paclitaxel conjugated to poly-(l-glutamic acid)

• ANG1005: Paclitaxel linked to angiopep-2 (brain peptide vector)

• Paccal®: a formulation using an excipient composed of a surfactant-based derivative of retinoic acid (XR-17) that results in a nanoparticle micellar preparation with high water solubility that eliminates the need for Cremophor®

(Khanna et al. J Vet Intern Med, 2015; 29(4): 1006-1012)

An example capsule formulation for derazantinib includes granules comprising the active ingredient together with a pregelatinized starch, lactose, crospovidone and magnesium stearate.

Kits

The invention also provides pharmaceutical products such as kits which may include a container with derazantinib and paclitaxel, which can be provided in amounts sufficient for treatment. The kits can thus include multiple containers that each include pharmaceutically effective amounts of the active ingredients. Optionally, instruments and/or devices necessary for administering the pharmaceutical composition(s) can also be included in the kits. Furthermore, the kits can include additional components, such as instructions or administration schedules, for treating a patient with cancer with the combinations of the invention. Accordingly, in a further aspect the invention provides a pharmaceutical product such as a kit for use in treating gastric adenocarcinoma, wherein derazantinib and paclitaxel are provided as separate dosage units. In some embodiments, the kit further comprises instructions for simultaneous, separate or sequential administration thereof for use in the treatment of gastric adenocarcinoma.

The invention also relates to a kit for stratifying patients with solid tumors (e.g. gastric adenocarcinoma) based on the level of TAMs. Accordingly, in a further aspect the invention provides a kit for selecting patients with a solid tumor (e.g. gastric adenocarcinoma) for treatment with derazantinib and paclitaxel, comprising reagents necessary for measuring the level of TAMs in a sample. Preferably, the reagents comprise a capture reagent comprising a detector for TAMs and a detector reagent.

In some aspects and embodiments a kit may comprise a comparator module which comprises a standard value to which the level of TAMs in the sample is compared. In a preferred embodiment, the comparator module is included in instructions for use of the kit. In another embodiment the comparator module is in the form of a display device, for example a strip of color or numerically coded material which is designed to be placed next to the readout of the sample measurement to indicate resistance levels. The standard value may be determined as described above.

The capture reagents may be antibodies or antibody fragments which selectively bind to biomarkers expressed on TAMs. For example, in some embodiments the reagents are in the form of one specific primary antibody which binds to TAMs and a secondary antibody which binds to the primary antibody, and which is itself labelled for detection. The primary antibody may also be labelled for direct detection. The kits or devices may optionally also contain a wash solution(s) that selectively allows retention of the bound biomarker to the capture reagent as compared with other biomarkers after washing. Such kits can then be used in ELISA, western blotting, flow cytometry (FACS), immunofluorescence microscopy, immunohistochemical or other immunochemical methods to detect the level of the biomarker. Nonantibody based specific probes may also be used to detect the level of TAMs.

In some embodiments the kits of the invention also include derazantinib and/or paclitaxel e.g. in a form suitable for administration to patients.

Additional Therapeutics

The combination of the invention may be used alone in the treatment of the medical conditions described herein. It is also contemplated that the combination is used together with a surgical procedure (for example to remove or reduce the size of a tumor), radiation therapy, ablation therapy and/or one or more therapeutic agents other than derazantinib and paclitaxel. Examples of anti-cancer agents that can be used together with the combination of the invention include but are not limited to chemotherapy (cytotoxic therapy), kinase inhibitors, endocrine therapy, biologies, immunotherapy, or a combination of these.

An additional therapeutic of particular interest is ramuciramab, marketed as Cyramza®. Ramuciramab is a vascular endothelial growth factor receptor 2 antagonist human monoclonal antibody. Ramuciramab is indicated for use as a single agent, or in combination with paclitaxel, indicated for the treatment of patients with advanced or metastatic, gastric or gastro-esophageal junction adenocarcinoma with disease progression on or after prior fluoropyrimidine -or platinum-containing chemotherapy. The recommended dose of ramuciramab (for gastric cancer) either as a single agent or in combination with weekly paclitaxel is 8 mg/kg every 2 weeks administered as an intravenous infusion over 60 minutes (FDA approved Label).

In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered once per day and paclitaxel and ramuciramab are administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15 and wherein ramuciramab is administered in weeks 1 and 3 of the 28-day cycle, e.g. on days 1 and 15.

In some embodiments, derazantinib is administered orally according to a continuous treatment schedule in which derazantinib is administered twice per day and paclitaxel and ramuciramab are administered by intravenous administration according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. on days 1, 8 and 15 and wherein ramuciramab is administered in weeks 1 and 3 of the 28-day cycle, e.g. on days 1 and 15.

When ramuciramab is used with the combination of the invention the dose may be less than the recommended dose, e.g. the dose may be about 5 mg/kg to about 8 mg/kg, e.g. about 5 mg/kg, about 6mg/kg, e.g. about 7 mg/kg, e.g. about 8 mg/kg.

When paclitaxel and ramuciramab are administered on the same day usually ramuciramab will be administered first with a delay of at least 60 minutes before administering paclitaxel.

The invention may be described by the following paragraphs: Pl . A pharmaceutical combination comprising derazantinib and paclitaxel for use in the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P2. The pharmaceutical combination for use according to paragraph Pl, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene encode an amino acid sequence that comprises at least one change compared to the amino acid sequence encoded by a reference FGFR1 DNA sequence, a reference FGFR2 DNA sequence and a reference FGFR3 DNA sequence respectively.

P3. The pharmaceutical combination for use according to paragraph P2, wherein the reference FGFR1 DNA sequence, the reference FGFR2 DNA sequence and the reference FGFR3 DNA sequence is the corresponding wildtype DNA sequence.

P4. The pharmaceutical combination for use according to paragraph P2, wherein the reference FGFR1 DNA sequence, the reference FGFR2 DNA sequence and the reference FGFR3 DNA sequence is a database of multiple corresponding DNA sequences in which sequences with known germline variations and somatic mutations from normal tissue are taken into account and not identified as a mutation.

P5. The pharmaceutical combination for use according to paragraph P2, wherein the reference FGFR1 DNA sequence is a DNA sequence encoding the amino acid sequence depicted in SEQ ID NO:2 or SEQ ID NO: 8, the reference FGFR2 DNA sequence is a DNA sequence encoding the amino acid sequence depicted in SEQ ID NO:4 or SEQ ID NO: 10 and the reference FGFR3 DNA sequence is a DNA sequence encoding the amino acid sequence depicted in SEQ ID NO:6 or SEQ ID NO: 12.

P6. The pharmaceutical combination for use according to any one of paragraphs Pl to P5, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene encode a functional tyrosine-kinase domain.

P7. The pharmaceutical combination for use according to paragraph P6, wherein the tyrosine kinase domain retains at least 20% of the phosphorylation activity compared to the tyrosine kinase domain encoded by the corresponding wildtype gene as measured in vitro by a protein kinase assay.

P8. The pharmaceutical combination for use according to any one of paragraphs Pl to P7, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene harbor a gene rearrangement, a splice site variant, and/or a short variant. P9. The pharmaceutical combination for use according to paragraph P8, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene harbor a gene rearrangement and the gene rearrangement is a gene fusion.

PIO. The pharmaceutical combination for use according to any one of paragraphs Pl to P9, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene do not harbor the following: i) a nonsense and/or frameshift mutation in exons 2-16 and in exon 17 up to amino acid 757 in FGFR1 (SEQ ID NO: 2), up to amino acid 761 in FGFR2 (SEQ ID NO: 4) and up to amino acid 751 in FGFR3 (SEQ ID NO:6); ii) a deletion in exons 11-17 encoding at least 30 consecutive amino acids: iii) a DNA sequence encoding a point amino acid substitution which inactivates tyrosine kinase activity; iv) a splice variant in introns 10-15; and v) a 5’ FGFR fusion with a breakpoint in introns 2-9.

Pl 1. The pharmaceutical combination for use according to any one of paragraphs Pl to P10, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene.

P12. The pharmaceutical combination for use according to any one of paragraphs Pl to P10, wherein the gastric adenocarcinoma harbors a mutated FGFR2 gene.

P13. The pharmaceutical combination for use according to any one of paragraphs Pl to P10, wherein the gastric adenocarcinoma harbors a mutated FGFR3 gene.

P14. The pharmaceutical combination for use according to any one of paragraphs Pl to P13, wherein derazantinib is administered orally.

P15. The pharmaceutical combination for use according to paragraph P14, wherein the daily oral dose is about lOOmg to about 400mg, e.g. about lOOmg to about 300mg.

P16. The pharmaceutical combination for use according to any one of paragraphs Pl to P15, wherein paclitaxel is administered according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week. P17. The pharmaceutical combination for use according to paragraph P16, wherein paclitaxel is administered intravenously and the weekly intravenous dosage amount of paclitaxel is about 60mg/m² to about 90mg/m², e.g. about 60mg/m² to about 80mg/m².

P18. The pharmaceutical combination for use according to any one of paragraphs Pl to P13, wherein derazantinib and paclitaxel are administered according to any one of embodiments 1 to 1116 in Table 1.

P19. The pharmaceutical combination for use according to any one of paragraphs Pl to P18, wherein derazantinib is administered orally and paclitaxel is administered intravenously and the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg is about 0.6: 1.0 to about 3.6: 1.0.

P20. The pharmaceutical combination for use according to any one of paragraphs Pl to P19, wherein the treatment additionally comprises administering ramuciramab to the patient.

P21. Derazantinib for use in combination with paclitaxel for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P22. Derazantinib for use according to paragraph P21, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13.

P23. Derazantinib for use according to paragraph P21 or paragraph P22, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

P24. Derazantinib for use according to any one of paragraphs P20 to P23, wherein the treatment additionally comprises administering ramuciramab to the patient.

P25. Paclitaxel for use in combination with derazantinib for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P26. Paclitaxel for use according to paragraph P25, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13.

P27. Paclitaxel for use according to paragraph P25 or paragraph P26, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19. P28. Paclitaxel for use according to any one of paragraphs P25 to P27, wherein the treatment additionally comprises administering ramuciramab to the patient.

P29. Use of derazantinib and paclitaxel in the manufacture of single agent medicaments for use in combination for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P30. Use of derazantinib and paclitaxel according to paragraph P29, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene is as defined in any one of paragraphs P2 to Pl 3.

P31. Use of derazantinib and paclitaxel according to paragraph P29 or paragraph P30, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

P32. Use of derazantinib and paclitaxel according to any one of paragraphs P29 to P31, wherein the treatment additionally comprises administering ramuciramab to the patient.

P33. Use of derazantinib in the manufacture of a medicament for use in combination with paclitaxel for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P34. Use of derazantinib according to paragraph P33, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13.

P35. Use of derazantinib according to paragraph P33 or paragraph P34, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

P36. Use of derazantinib according to any one of paragraphs P33 to P35, wherein the treatment additionally comprises administering ramuciramab to the patient.

P37. Use of paclitaxel in the manufacture of a medicament for use in combination with derazantinib for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P38. Use of paclitaxel according to paragraph P37, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13. P39. Use of paclitaxel according to paragraph P37 or paragraph P38, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

P40. Use of paclitaxel according to any one of paragraphs P37 to P39, wherein the treatment additionally comprises administering ramuciramab to the patient.

P41. A method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of derazantinib and a therapeutically effective amount of paclitaxel, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P42. The method according paragraph P41, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13.

P43. The method according to paragraph P41 or paragraph P42, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

P44. The method according to any one of paragraphs P41 to P43, wherein the treatment additionally comprises administering ramuciramab to the patient.

P45. A method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of derazantinib, which patient is undergoing or will undergo treatment with paclitaxel, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P46. The method according to paragraph P45, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13.

P47. The method according to paragraph P45 or paragraph P46, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

P48. The method according to any one of paragraphs P45 to P47, wherein the treatment additionally comprises administering ramuciramab to the patient.

P49. A method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of paclitaxel, which patient is undergoing or will undergo treatment with derazantinib, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

P50. The method according to paragraph P49, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13.

P51. The method according to paragraph P49 or paragraph P50, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

P52. The method according to any one of paragraphs P49 to P51, wherein the treatment additionally comprises administering ramuciramab to the patient.

P53. The pharmaceutical combination for use according to any one of paragraphs Pl to P20, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value.

P54. The pharmaceutical combination for use according to paragraph P53, wherein a higher level of TAMs in a tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

P55. The pharmaceutical combination for use according to paragraph P53, when the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

P56. Derazantinib for use according to any one of paragraphs P21 to P24, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value.

P57. Derazantinib for use according to paragraph P56, wherein a higher level of TAMs in a tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid. P58. Derazantinib for use according to paragraph P56, when the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

P59. Paclitaxel for use according to any one of paragraphs P25 to P28, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value.

P60. Paclitaxel for use according to paragraph P59, wherein a higher level of TAMs in a tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

P61. Paclitaxel for use according to paragraph P59, when the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

P62. Use of derazantinib and paclitaxel according to any one of paragraphs P29 to P32, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value.

P63. Use of derazantinib and paclitaxel according to paragraph P62, wherein a higher level of TAMs in a tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

P64. Use of derazantinib and paclitaxel according to paragraph P62, when the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

P65. Use of derazantinib according to any one of paragraphs P33 to P36, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value. P66. Use of derazantinib according to paragraph P65, wherein a higher level of TAMs in a tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

P67. Use of derazantinib according to paragraph P65, when the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

P68. Use of paclitaxel according to any one of paragraphs P37 to P40, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value.

P69. Use of paclitaxel according to paragraph P68, wherein a higher level of TAMs in a tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

P70. Use of paclitaxel according to paragraph P68, when the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

P71. The method according to any one of paragraphs P41 to P52, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value.

P72. The method according to paragraph P71, wherein a higher level of TAMs in a tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

P73. The method according to paragraph P71 , when the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample. QI. A pharmaceutical combination comprising derazantinib and paclitaxel for use in the treatment of a patient with a solid tumor, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

Q2. The pharmaceutical combination for use according to paragraph QI, wherein the solid tumor is gastric adenocarcinoma.

Q3. The pharmaceutical combination for use according to paragraph QI or Q2, wherein a higher level of TAMs in the tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

Q4. The pharmaceutical combination for use according to paragraph QI or Q2, wherein a higher level of TAMs in a tumor sample is determined relative to a standard value from subjects with the same tumor histotype.

Q5. The pharmaceutical combination for use according to paragraph QI or Q2, wherein the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

Q6. The pharmaceutical combination for use according to any one of paragraphs QI to Q5, wherein the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

Q7. The pharmaceutical combination for use according to paragraph Q6, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and/or the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13. Q8. The pharmaceutical combination for use according to any one of paragraphs QI to Q7, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

Q9. The pharmaceutical combination for use according to paragraph Q8, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a copy number of 2 (mean average).

Q10. The pharmaceutical combination for use according to paragraph Q8, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a copy number of 6 (mean average).

Q 11. The pharmaceutical combination for use according to any one of paragraphs Q 1 to Q 10, wherein the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

Q12. The pharmaceutical combination for use according to paragraph QI 1, wherein the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is at least about 50% greater than the matched-tissue non-tumor FGFR1 gene.

Q 13. The pharmaceutical combination for use according to any one of paragraphs Q 1 to Q 12, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

Q 14. The pharmaceutical combination for use according to any one of paragraphs Q 1 to Q 13, wherein the treatment additionally comprises administering ramuciramab to the patient.

QA 1. Derazantinib for use in combination with paclitaxel for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

(iii) the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value. QA2. Derazantinib for use according to paragraph QA1, wherein the solid tumor is gastric adenocarcinoma.

QA3. Derazantinib for use according to paragraph QA1 or QA2, wherein a higher level of TAMs in the tumor sample is determined as defined in any one of paragraphs Q3 to Q5.

QA4. Derazantinib for use according to any one of paragraphs QA1 to QA3, wherein the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene as defined in paragraphs Q6 or Q7; and/or the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in any one of paragraphs Q8 to Q10; and/or the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in paragraph Q 11 or Q 12.

QA5. Derazantinib for use according to any one of paragraphs QA1 to QA4, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

QA6. Derazantinib for use according to any one of paragraphs QA1 to QA5, wherein the treatment additionally comprises administering ramuciramab to the patient.

QB 1. Paclitaxel for use in combination with derazantinib for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

QB2. Paclitaxel for use according to paragraph QB1, wherein the solid tumor is gastric adenocarcinoma.

QB3. Paclitaxel for use according to paragraph QB 1 or QB2, wherein a higher level of TAMs in the tumor sample is determined as defined in any one of paragraphs Q3 to Q5. QB4. Paclitaxel for use according to any one of paragraphs QB 1 to QB3, wherein the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene as defined in paragraphs Q6 or Q7; and/or the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in any one of paragraphs Q8 to Q10; and/or the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in paragraph Q 11 or Q 12.

QB5. Paclitaxel for use according to any one of paragraphs QB 1 to QB4, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

QB6. Paclitaxel for use according to any one of paragraphs QB 1 to QB5, wherein the treatment additionally comprises administering ramuciramab to the patient.

QC1. Use of derazantinib and paclitaxel in the manufacture of single agent medicaments for use in combination for the treatment of a patient with a solid tumor, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

The embodiments described in Paragraphs Q2 to Q14 apply similarly to paragraph QC1.

QD 1. Use of derazantinib in the manufacture of a medicament for use in combination with paclitaxel for the treatment of a patient with a solid tumor, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

(iii) the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value. The embodiments described in Paragraphs QA2 to QA6 apply similarly to paragraph QD1.

QE1. Use of paclitaxel in the manufacture of a medicament for use in combination with derazantinib for the treatment of a patient with a solid tumor, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

The embodiments described in Paragraphs QB2 to QB6 apply similarly to paragraph QE1.

QF1. A method of treating a patient with a solid tumor comprising administering to the patient a therapeutically effective amount of derazantinib and a therapeutically effective amount of paclitaxel, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

QF2. The method according to paragraph QF1, wherein the solid tumor is gastric adenocarcinoma.

QF3. The method according to paragraph QF1 or QF2, wherein a higher level of TAMs in the tumor sample is determined as defined in any one of paragraphs Q3 to Q5.

QF4. The method according to any one of paragraphs QF1 to QF3, wherein the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene as defined in paragraphs Q6 or Q7; and/or the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in any one of paragraphs Q8 to Q10; and/or the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in paragraph Q 11 or Q 12.

QF5. The method according to any one of paragraphs QF1 to QF4, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

QF6. The method according to any one of paragraphs QF1 to QF5, wherein the treatment additionally comprises administering ramuciramab to the patient.

QG1. A method of treating a patient with a solid tumor comprising administering to the patient a therapeutically effective amount of derazantinib, which patient is undergoing or will undergo treatment with paclitaxel, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the gastric adenocarcinoma has at least one of the following characteristics:

QG2. The method according to paragraph QG1, wherein the solid tumor is gastric adenocarcinoma.

QG3. The method according to paragraph QG1 or QG2, wherein a higher level of TAMs in the tumor sample is determined as defined in any one of paragraphs Q3 to Q5.

QG4. The method according to any one of paragraphs QG1 to QG3, wherein the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene as defined in paragraphs Q6 or Q7; and/or the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in any one of paragraphs Q8 to Q10; and/or the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in paragraph Q 11 or Q 12.

QG5. The method according to any one of paragraphs QG1 to QG4, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19. QG6. The method according to any one of paragraphs QG1 to QG5, wherein the treatment additionally comprises administering ramuciramab to the patient.

QH1. A method of treating a patient with a solid tumor comprising administering to the patient a therapeutically effective amount of paclitaxel, which patient is undergoing or will undergo treatment with derazantinib, wherein the gastric adenocarcinoma has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the gastric adenocarcinoma has at least one of the following characteristics:

QH2. The method according to paragraph QH1, wherein the solid tumor is gastric adenocarcinoma.

QH3. The method according to paragraph QH1 or QH2, wherein a higher level of TAMs in the tumor sample is determined as defined in any one of paragraphs Q3 to Q5.

QH4. The method according to any one of paragraphs QH1 to QH3, wherein the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene as defined in paragraphs Q6 or Q7; and/or the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in any one of paragraphs Q8 to Q10; and/or the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value as defined in paragraph Q 11 or Q 12.

QH5. The method according to any one of paragraphs QH1 to QH4, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

QH6. The method according to any one of paragraphs QH1 to QH5, wherein the treatment additionally comprises administering ramuciramab to the patient.

R1. A method of selecting a patient with a solid tumor for treatment with derazantinib and paclitaxel, wherein the solid tumor has at least one of the following characteristics: (i) the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene;

R2. The method according to paragraph Rl, wherein the solid tumor is gastric adenocarcinoma.

R3. The method according to paragraph Rl or R2, wherein a higher level of TAMs in the tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

R4. The method according to paragraph Rl or R2, wherein a higher level of TAMs in a tumor sample is determined relative to a standard value from subjects with the same tumor histotype.

R5. The method according to paragraph Rl or R2, wherein the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

R6. The method according to any one of paragraphs Rl to R5, wherein the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

R7. The method according to paragraph R6, wherein the method includes prior to step (b): determining ex vivo in a sample taken from the patient whether the patient’s solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

R8. The method according to paragraph R6 or R7, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and/or the mutated FGFR3 gene is as defined in any one of paragraphs P2 to P13. R9. The method according to any one of paragraphs R1 to R8, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

R10. The method according to paragraph R9, wherein the method includes prior to step (b): determining ex vivo in a sample taken from the patient whether the patient’s solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

R11. The method according to paragraph R9 or R10, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a copy number of 2 (mean average).

R12. The method according to paragraph R9 or R10, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a copy number of 6 (mean average).

R13. The method according to any one of paragraphs R1 to R12, wherein the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

R14. The method according to paragraphs R13, wherein the method comprises prior to step (b), determining ex vivo in a sample taken from the patient whether the patient’s solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

R15. The method according to paragraph R13 or R14, wherein the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is at least about 50% greater than the matched-tissue non-tumor FGFR1 gene.

R16. The method according to any one of paragraphs R1 to R15, wherein the method comprises after step (b), (cl) treating the patient with a therapeutically effective amount of derazantinib and paclitaxel.

R17. The method according to any one of paragraphs R1 to R15, wherein the method comprises after step (b), (c2) treating the patient with a therapeutically effective amount of derazantinib, which patient is undergoing or will undergo treatment with paclitaxel. R18. The method according to any one of paragraphs R1 to R15, wherein the method comprises after step (b), (c3) treating the patient with a therapeutically effective amount of paclitaxel, which patient is undergoing or will undergo treatment with derazantinib .

R19. The method according to any one of paragraphs R16 to R18, wherein derazantinib and/or paclitaxel is administered as defined in any one of paragraphs P14 to P19.

R20. The method according to any one of paragraphs R1 to R19, wherein the method additionally comprises administering ramuciramab to the patient.

All amino acid sequences are depicted with the N-terminus on the left and C-terminus on the right. The quantification of gaps or insertions in DNA and amino acid sequences is determined using the National Institute of Health Basica Local Alignment Search Tool (BLAST) using the default parameters, available at https://blast.ncbi.nlm.nih.gov/Blast.cgi.

All aspects and embodiments of the invention described herein may be combined in any combination where possible.

A number of publications are cited herein in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Each of these references is incorporated herein by reference in its entirety into the present disclosure, to the same extent as if each individual reference was specifically and individually indicated to be incorporated by reference.

Particular embodiments of the invention are described in the following Examples, which serve to illustrate the invention in more detail and should not be construed as limiting the invention in any way.

Examples

Female Balb/c nude mice (CrownBio) of at least 20 g body-weight were subcutaneously (s.c.) inoculated with tumor fragments (3*3*3 mm) from stock mice bearing the respective s.c. patient-derived-xenograft (PDX) tumors. For the cell-line derived xenograft (CDX) model, SNU-16, cultured cells were injected s.c. at a dose of 10⁷ cells (in 0. 1 mL) per mouse. Randomization into different treatment groups (5-9 mice per group) was made when the mean tumor size was 150-200 mm³. For the PDX-experiments this was always n=5/group, which was a relatively small number per group, but allowed a screening approach to determine if there were any important signals of a positive interaction of the two test compounds; with such relatively low numbers, the statistical power would be low. Mice were dosed daily with derazantinib (35 mg/kg, p.o.) or vehicle (5 mL/kg of a solution of DMA:Cremophor-EL:Propylene-Glycol:0.2M acetate buffer at pH5 in the ratios of 10: 10:30:50), or weekly with commercially available paclitaxel (15 mg/kg, i.v. as a 20-30 sec infusion of 5 mL/kg), or as a combination in which the weekly dose of paclitaxel was administered 30 min after administration of derazantinib. Paclitaxel was purchased from Peking Union Medical College as a solution (6 mg/mL in anhydrous citrate, polyoxyethylated (35) castor oil and absolute ethyl alcohol). Derazantinib was supplied as a powder by Basilea Pharmaceutica International Ltd. and was prepared as a solution as above and stored for 7 days at 4°C.

Body weights were determined daily and tumor volumes were determined at least twice per week by measuring two dimensions with calipers and applying the formula “V = (L x W²)/2”, where V is the tumor volume, and L and W are the tumor length and width respectively. Individual mice were culled when tumors reached 1500 mm³ or more, or when the body-weight loss (BWL) was found to have exceeded 20%. Mice were also culled if the BWL was determined to be >15% for 3 consecutive days. Any mice with >10% BWL, automatically received a dosing -holiday until the BWL returned to <10%. All animal protocols were reviewed and approved by the relevant local committees in China, which is where the studies were performed.

The statistical significance of effects were determined at the endpoint, which was after 4-weeks treatment or when the tumors reached 1500 mm³ in size. For efficacy, a delta treated/control ratio was determined (dT/C) i.e. where the change in tumor volume for the treated-group was divided by the change in tumorvolume for the vehicle -control group; thus the lower that value, the greater the effect, and a negative value would indicate regression. A one-way ANOVA was performed with Holms-Sidak applied post-hoc for multiple determinations, where a p-value of <0.05 was considered statistically significant. In addition, an assessment of synergy was made based upon the “Clarke-Index” to estimate the degree of synergy: T/C B - (T/CA*T/CB) in which A and B are two different compounds, where CI < -0.1 indicates synergy (or a positive interaction), +0. 1 (antagonism, or a negative interaction) and between -0. 1 and +0.1 indicates additivity (no interaction); see O’Reilly et al. (Anti Cancer Drugs 2011;22:58-78). These analyses are summarized in Table 2.

The M2 -macrophage content of each tumor was quantified by immonhistochemistry (IHC) of formalin- fixed paraffin-embedded (FFPE) tumor slices of 4 pm, which were prepared from vehicle-treated tumors at the end of an efficacy experiment. All stained sections were scanned with the NanoZoomer-HT 2.0 Image system (Hamamatsu Photonics) and Pannoramic Digital Slide Scanners (3DHISTECH, Pannoramic SCAN). A high resolution picture for the whole section was generated and further analyzed. M2-macrophages were identified by Arginase- 1 staining and the data expressed as the percentage of all cells present in the tumor. Table 2. Summary of endpoint-analyses for efficacy.

The p-value is for the comparison of all groups in that model (one-way analysis of variance with Holms- Sidak, post-hoc). The Clarke-combination-index (CCI) indicates synergy (<-0.1) or antagonism (>0.1) or additive if neither (O’Reilly et al.).

In the GA3055 PDX-model (FGFR2 -fusion (LINCO1153)) all treatments caused a highly significant inhibition of tumor growth compared to the vehicle-treated group (p<0.001). Moreover, the combination induced regression in 5/5 cases, 3 of which led to complete regression by the endpoint (no tumors visible). Indeed, at the endpoint, the combination was significantly different to both the paclitaxel-group (p=0.02) and the derazantinib-group (p=0.03). However, the combination was less well tolerated, and by the endpoint, it had caused significantly greater body-weight loss (p<0.002) than any of the other treated groups, although the mean decrease in body-weight was relatively small at -2±1. 1%. Thus overall, the combination significantly increased efficacy, which was well tolerated by the mice (Figure 1AB).

In the GA6208 PDX-model (two different FGFR genetic-aberrations: an FGFR2 mutation (Y244C) and an FGFR3 mutation (G224S)), neither monotherapy had any efficacy at all, but in contrast, the combination showed a strong inhibition of growth which was highly significantly different to all other treatments (p<0.01). Furthermore, during the final week of treatment, 2/5 tumors began to show regression (-9% and -28%), while the other tumors showed little or no growth to give an overall dT/C of 0.1 at the endpoint of day-27. The growth of this tumor did not appear to be well tolerated by the mice, since there was a gradual decrease in body-weight in the vehicle control group during the 4-weeks of treatment reaching a mean±SEM of -4±3% at the endpoint. Consequently, any treatment tended to make the body-weight loss slightly worse, although this did not reach significance (p=0.7). Thus, overall, the combination strongly and significantly increased efficacy, which did not significantly increase bodyweight loss. In the repeat combination-experiment of this model, strong synergy was again seen (CCI= - 0.42), although less than in the first experiment, because of more activity from the paclitaxelmonotherapy (Figure 2ABCD).

In the GA3236 PDX-model (wild type, FGFR3 moderately increased expression), none of the treatments caused a significant decrease in tumor growth (p=0.12), although both monotherapies had a lower dT/C (0.59) than the vehicle, and the combination had a dT/C of 0.33 on day-21, at which point it was necessary to start culling mice with tumors >1500 mm³. An exponential growth-fit of the individual mice also failed to show a significant difference between any of the treatment groups (p=0.3; results not shown). Overall, the responses in the combination-group were very variable, with one tumor showing strong regression from day-14, which reached 90%-regression two weeks later. In contrast, the other 4 tumors grew similarly to the two monotherapy treatments, and thus there was overall no significant difference. The CCI value of -0.02 was consistent with an overall additive effect of the combination. The tolerability of all the treatments tended to reduce body-weight compared to the vehicle-group (Figure 3B), although this did not reach significance (p=0.39).

In the GA0095 PDX-model (wild type, FGFR3 moderately increased expression), both monotherapies caused slight, non-significant, growth inhibition (dT/Cs of 0.6), but the combination significantly inhibited growth compared to the vehicle (p<0.01) with a low dT/C of 0.19, although none of the 5 tumors showed regression (Figure 4A). The tumor appeared to be poorly tolerated since there was marked body-weight loss in the vehicle-treated group of ca. 5%. None of the treatments made the body-weight loss significantly worse (p=0.43); indeed, the combination seemed overall best-tolerated, probably because it reduced the tumor volume (Figure 4B).

In the GAO 114 PDX-model (wild type, FGFR2 highly amplified and moderately increased expression), derazantinib was without any efficacy at all, while paclitaxel caused significant regression of -27±40% (mean±SEM) by the endpoint of day-27 (Figure 5A). Combination of the two monotherapies did not improve efficacy but actually slightly reduced it, although this did not reach significance, and overall the CCI was +0.1, a value consistent with an additive effect tending towards antagonistic. The combination did however increase the body-weight loss and this was significant at the endpoint (Figure 5B).

In the SNU-16 CDX-model (FGFR2 -fusion (PDHX)), derazantinib inhibited tumor growth, but due to the variability within the vehicle-treated group, this did not reach significance (Figure 6A). Paclitaxel however, caused strong regression by day-27 of -82±4%, which included one complete response (100% regression) out of the 8 mice treated. The combination gave an even stronger response with 8/8 tumors showing complete-regression by day-27. The paclitaxel treatment was well-tolerated, but the derazantinib monotherapy caused marked body-weight loss, although this was not increased in the combination (Figure 6B). In the repeat combination-experiment of this model, paclitaxel again induced strong regression, mean of -44±14% (dT/C= -0.62) and 6/7 tumors regressed. DZB high-dose monotherapy (75 mg/kg) was without statistically significant efficacy (dT/C= 0.3), even though there were 3/7 tumors that had regressed by day-26, and DZB low-dose monotherapy was completely without activity (dT/C=1.26). In combination with paclitaxel, DZB (35 mg/kg) was highly efficacious causing all 7/7 tumors to regress by day-26, and although there were no complete regressions, the mean overall regression was -65 ±4% by the endpoint of day-26. As a further assessment, the CCI was calculated to be -0.16, which was again consistent with synergy.

In the GA0031 PDX-model (FGFR1 mutation (R285W (R254W with respect to SEQ ID NO: 2)), DZB low-dose monotherapy was completely without activity (dT/C=1.13), and both paclitaxel and DZB high- dose therapy had only weak effects (dT/Cs of 0.69 and 0.71). The lowest dT/C was for the combination (dT/C= 0.5), which however did not reach significance, although the calculated CCI was -0.28, which was consistent with synergy. The combination showed a trend to increase body-weight loss, but this was not significant, while the high-dose DZB-group did cause significant body-weight loss compared to the vehicle group.

Thus, the combination of derazantinib and paclitaxel in treating subcutaneously-grown PDX- or CDX- models, each with a different type of FGFR-aberration, including fusion, DNA mutation encoding an amino acid change, amplification and/or increased FGFR expression, led in two cases to additive efficacy and synergy in the other 5 cases. One case resulted in remarkable synergy, although the degree of synergy was found to be slightly less on repeating the experiment due to higher activity of the paclitaxel monotherapy (Figure 2CD, Table 2).

It is clear that other important signaling pathways may be activated in a different manner to a different extent in the various tumor models, but also the individual tumor microenvironment could play an important role. Immunohistochemical (IHC) analysis of the vehicle-treated tumors for blood-vessel density or maturity (CD31 and SMA, respectively) or proliferation (Ki67), did not reveal any further insights into why one model would be more sensitive to the combination than another. However, the percentage of cells positive for M2 or Ml -type tumor-associated-macrophages, i.e. TAMs (determined by IHC) did reveal an interesting pattern. M2 -TAMS express the CSF1R which is important for their function, and this receptor is also a target for derazantinib. The percentage of Ml -TAMs was very low (median of 0.7%) and rather invariant, while M2s were more abundant (median of 2.2%) and varied from 0.7-8.7% across the 7 different gastric-models. The two models that showed an at best additive -effect (GA3236 PDX-model, FGFR3 wild type, not amplified with moderately increased expression, GAO 114 PDX-model, FGFR2 wild type highly amplified, but with moderately increased expression), had a median %M2-TAM of <0.7, while all the synergistic models were >1.2% (Figure 8, Table 3); which was a significant difference (p=0.02, Fisher’s 1-tailed exact-test). This suggests that tumors with FGFR mutations, or high FGFR gene expression or gene amplification, as well as higher levels of M2 -TAM infiltration, may be more sensitive to the derazantinib-paclitaxel combination. Such a concept is consistent with the mechanism of action of derazantinib which, based upon kinase assays, inhibits FGFR1-3 and CSF1R with similar potency (McSheehy et al, Mol Cancer Ther 2019; 18( 12 Suppl)).

Table 3. Gastric models and their FGFR-aberrations with %M2-TAM content

*M2-TAMs were quantified by IHC of cells staining positive for Arginase- 1 as described in the Methods above.

**The levels of RNA expression (shown as a Log2 value) were determined by RNA-Seq for each model and were provided by the company CrownBio Inc.

*** From repeat experiments (GA6208 and SNU-16).

In conclusion, derazantinib combined with paclitaxel leads to significantly increased efficacy, which in some models is additive, but in others synergistic, and this synergy may be partially influenced by the level of infiltration of the tumor by M2-TAMs. Overall, the selected doses and schedules used in the combination tended to slightly increase body-weight loss compared to the other treatment -groups, but on average this was never greater than 5% indicating that the combination was relatively well -tolerated by the mice over the four week period of the experiment.

Claims

77 Claims

1. A pharmaceutical combination comprising derazantinib and paclitaxel for use in the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

2. The pharmaceutical combination for use according to claim 1, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene encode an amino acid sequence that comprises at least one change compared to the amino acid sequence encoded by a reference FGFR1 DNA sequence, a reference FGFR2 DNA sequence and a reference FGFR3 DNA sequence respectively.

3. The pharmaceutical combination for use according to claim 2, wherein the reference FGFR1 DNA sequence, the reference FGFR2 DNA sequence and the reference FGFR3 DNA sequence is the corresponding wildtype DNA sequence.

4. The pharmaceutical combination for use according to claim 2, wherein the reference FGFR1 DNA sequence, the reference FGFR2 DNA sequence and the reference FGFR3 DNA sequence is a database of multiple corresponding DNA sequences in which sequences with known germline variations and somatic mutations from normal tissue are taken into account and not identified as a tumor-associated mutation.

5. The pharmaceutical combination for use according to claim 2, wherein the reference FGFR1 DNA sequence is a DNA sequence encoding the amino acid sequence depicted in SEQ ID NO:2 or SEQ ID NO: 8, the reference FGFR2 DNA sequence is a DNA sequence encoding the amino acid sequence depicted in SEQ ID NO:4 or SEQ ID NO: 10 and the reference FGFR3 DNA sequence is a DNA sequence encoding the amino acid sequence depicted in SEQ ID NO:6 or SEQ ID NO: 12.

6. The pharmaceutical combination for use according to any one of claims 1 to 5, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene encode a functional tyrosine-kinase domain.

7. The pharmaceutical combination for use according to claim 6, wherein the tyrosine kinase domain retains at least 20% of the phosphorylation activity compared to the tyrosine kinase domain encoded by the corresponding wildtype gene as measured in vitro by a protein kinase assay. 78

8. The pharmaceutical combination for use according to any one of claims 1 to 7, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene harbor a gene rearrangement, a splice site variant, and/or a short variant.

9. The pharmaceutical combination for use according to claim 8, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene harbor a gene rearrangement and the gene rearrangement is a gene fusion.

10. The pharmaceutical combination for use according to any one of claims 1 to 9, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and the mutated FGFR3 gene do not harbor the following: i) a nonsense and/or frameshift mutation in exons 2-16 and in exon 17 up to amino acid 757 in FGFR1 (SEQ ID NO: 2), up to amino acid 761 in FGFR2 (SEQ ID NO: 4) and up to amino acid 751 in FGFR3 (SEQ ID NO: 6); ii) a deletion in exons 11-17 encoding at least 30 consecutive amino acids: iii) a DNA sequence encoding a point amino acid substitution which inactivates tyrosine kinase activity; iv) a splice variant in introns 10-15; and v) a 5’ FGFR fusion with a breakpoint in introns 2-9.

11. The pharmaceutical combination for use according to any one of claims 1 to 10, wherein derazantinib is administered orally, e.g. wherein the daily oral dose is about lOOmg to about 400mg, e.g. about lOOmg to about 300mg; and/or wherein paclitaxel is administered according to a cyclic treatment schedule wherein the treatment cycle is 28 days and paclitaxel is administered once per week for three weeks followed by a rest week, e.g. wherein paclitaxel is administered intravenously and the weekly intravenous dosage amount of paclitaxel is about 60mg/m² to about 90mg/m² , e.g. about 60mg/m² to about 80mg/m², e.g. wherein derazantinib and paclitaxel are administered according to any one of embodiments 1 to 1116 in Table 1.

12. The pharmaceutical combination for use according to any one of claims 1 to 11, wherein derazantinib is administered orally and paclitaxel is administered intravenously and the mass ratio of the daily dose of derazantinib to the weekly intravenous dose of paclitaxel in terms of mg/kg is about 0.6: 1.0 to about 3.6: 1.0.

13. The pharmaceutical combination for use according to any one of claims 1 to 12, wherein the treatment additionally comprises administering ramuciramab to the patient. 79

14. The pharmaceutical combination for use according to any one of claim 1 to 13, wherein the gastric adenocarcinoma has a level of TAMs determined to be higher than a standard value.

15. The pharmaceutical combination for use according to claim 14, wherein a higher level of TAMs in a tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

16. The pharmaceutical combination for use according to claim 14, when the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

17. Derazantinib for use in combination with paclitaxel for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

18. Paclitaxel for use in combination with derazantinib for the treatment of a patient with gastric adenocarcinoma harboring a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

19. A method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of derazantinib and a therapeutically effective amount of paclitaxel, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

20. A method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of derazantinib, which patient is undergoing or will undergo treatment with paclitaxel, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

21. A method of treating a patient with gastric adenocarcinoma comprising administering to the patient a therapeutically effective amount of paclitaxel, which patient is undergoing or will undergo treatment with derazantinib, wherein the gastric adenocarcinoma harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

22. A pharmaceutical combination comprising derazantinib and paclitaxel for use in the treatment of 80 a patient with a solid tumor, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

23. The pharmaceutical combination for use according to claim 22, wherein the solid tumor is gastric adenocarcinoma.

24. The pharmaceutical combination for use according to claim 22 or claim 23, wherein a higher level of TAMs in the tumor sample is determined: i) relative to a standard value from subjects with the same tumor histotype; ii) relative to the level of TAMs in a sample taken from the same subject after treatment initiation; or iii) relative to a standard value from normal cells, tissue or body fluid.

25. The pharmaceutical combination for use according to claim 22 or claim 23, wherein a higher level of TAMs in a tumor sample is determined relative to a standard value from subjects with the same tumor histotype.

26. The pharmaceutical combination for use according to claim 22 or claim 23, wherein the level of TAMs is determined to be higher than a standard value when the level of TAM cells in a sample is a mean average of at least 1% of the total number of cells in a tumor sample.

27. The pharmaceutical combination for use according to any one of claims 22 to 26, wherein the solid tumor harbors a mutated FGFR1 gene, a mutated FGFR2 gene and/or a mutated FGFR3 gene.

28. The pharmaceutical combination for use according to claim 27, wherein the mutated FGFR1 gene, the mutated FGFR2 gene and/or the mutated FGFR3 gene is as defined in any one of claims 2 to 10.

29. The pharmaceutical combination for use according to any one of claims 22 to 28, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value. 81

30. The pharmaceutical combination for use according to claim 29, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a copy number of 2 (mean average).

31. The pharmaceutical combination for use according to claim 29, wherein the solid tumor has a level of gene amplification of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a copy number of 6 (mean average).

32. The pharmaceutical combination for use according to any one of claims 22 to 31, wherein the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is higher than a standard value.

33. The pharmaceutical combination for use according to claim 32, wherein the solid tumor has a level of expression (e.g. mRNA expression) of the FGFR1 gene, the FGFR2 gene and/or the FGFR3 gene which is at least about 50% greater than the respective matched-tissue non-tumor FGFR gene.

34. The pharmaceutical combination for use according to any one of claims 22 to 33, wherein derazantinib and/or paclitaxel is administered as defined in claim 11 or claim 12.

35. The pharmaceutical combination for use according to any one of claims 22 to 34, wherein the treatment additionally comprises administering ramuciramab to the patient.

36. Derazantinib for use in combination with paclitaxel for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

37. Paclitaxel for use in combination with derazantinib for the treatment of a patient with a solid tumor (e.g. gastric adenocarcinoma), wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics: 82

38. A method of treating a patient with a solid tumor (e.g. gastric adenocarcinoma) comprising administering to the patient a therapeutically effective amount of derazantinib and a therapeutically effective amount of paclitaxel, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

39. A method of treating a patient with a solid tumor (e.g. gastric adenocarcinoma) comprising administering to the patient a therapeutically effective amount of derazantinib, which patient is undergoing or will undergo treatment with paclitaxel, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics:

40. A method of treating a patient with a solid tumor (e.g. gastric adenocarcinoma) comprising administering to the patient a therapeutically effective amount of paclitaxel, which patient is undergoing or will undergo treatment with derazantinib, wherein the solid tumor has a level of TAMs determined ex vivo in a sample taken from the patient to be higher than a standard value and wherein the solid tumor has at least one of the following characteristics: 83

41. A method of selecting a patient with a solid tumor for treatment with derazantinib and paclitaxel, wherein the solid tumor has at least one of the following characteristics: