Genetic Alterations to the Selective FGFR Inhibitor Erdafitinib

Published OnlineFirst April 17, 2017; DOI: 10.1158/1535-7163.MCT-16-0518

Companion Diagnostics and Cancer Biomarkers Molecular Cancer Therapeutics Oncogenic Characterization and Pharmacologic Sensitivity of Activating Fibroblast Growth Factor Receptor (FGFR) Genetic Alterations to the Selective FGFR Inhibitor Erdaﬁtinib Jayaprakash D. Karkera1, Gabriela Martinez Cardona1, Katherine Bell1, Dana Gaffney1, Joseph C. Portale1, Ademi Santiago-Walker1, Christopher H. Moy1, Peter King1, Michael Sharp1, Rastislav Bahleda2, Feng R. Luo3, John D. Alvarez1, Matthew V.Lorenzi1, and Suso J. Platero1

Abstract

Fibroblast growth factor receptor (FGFR) genetic alterations are amplifications present in clinical specimens were shown to frequently observed in cancer, suggesting that FGFR inhibition display potent transforming activity associated with constitu- may be a promising therapy in patients harboring these lesions. tive pathway activation. Tumor cells expressing these FGFR Identification of predictive and pharmacodynamic biomarkers to activating mutants displayed sensitivity to the selective FGFR select and monitor patients most likely to respond to FGFR inhibitor erdafitinib and resulted in suppression of FGFR phos- inhibition will be the key to clinical development of this class phorylation and downstream signal transduction. Clinically, of agents. Sensitivity to FGFR inhibition and correlation with patients receiving erdafitinib showed decreased Erk phosphor- FGFR pathway activation status were determined in molecularly ylation in tumor biopsies and elevation of serum phosphate. In annotated panels of cancer cell lines and xenograft models. a phase I study, a heavily pretreated bladder cancer patient with Pathway inhibition in response to FGFR inhibitor treatment was an FGFR3–TACC3 translocation experienced a partial response assessed in cell lines (both in vitro and in vivo) and in samples from when treated with erdafitinib. This preclinical study confirmed patients treated with the FGFR inhibitor JNJ-42756493 (erdafiti- pharmacodynamics and identified new predictive biomarkers nib). Frequency of FGFR aberrations was assessed in a panel of to FGFR inhibition with erdafitinib and supports further clinical NSCLC, breast, prostate, ovarian, colorectal, and melanoma evaluation of this compound in patients with FGFR genetic human tumor tissue samples. FGFR translocations and gene alterations. Mol Cancer Ther; 16(8); 1717–26. 2017 AACR.

Introduction growth factor receptor (EGFR)-2 amplification, which predicts positive response to trastuzumab in women with breast cancer The successful development of new oncology therapies most (1), and KRAS mutation, which predicts resistance to the EGFR unequivocally requires identifying the patients most likely to inhibitors panitumumab and cetuximab in individuals with respond to a particular targeted intervention. Thus, the discovery colorectal tumors (2, 3). In lung cancer, EGFR-activating point and characterization of biomarkers that accurately distinguish mutations predict activity to EGFR tyrosine kinase inhibitors those more likely to respond patients from a broader population erlotinib and gefitinib (4), and ALK translocation has been shown is critical for successful clinical development. Examples of such to be a predictor of response to crizotinib, a compound targeting biomarkers that have already been identified and approved as the ALK tyrosine kinase (5). More recently, fibroblast growth predictors of response to therapy include human epidermal factor receptor (FGFR)-mediated signaling has been shown to play a role in tumor progression, and recent evidence suggests that FGFRs are oncogenic drivers in various types of cancer (6), 1Janssen Research and Development, LLC, Spring House, Pennsylvania. 2Drug highlighting the need to find predictive biomarkers of FGFR to Development Department (DITEP), Gustave Roussy Cancer Campus and Uni- support clinical development. 3 versity Paris-Sud, Villejuif, France. Janssen Research and Development, LLC, Fibroblast growth factors (FGF) are a family of secreted factors Raritan, New Jersey. involved in signaling pathways responsible for embryonic devel- Note: Supplementary data for this article are available at Molecular Cancer opment, cell proliferation, survival, and migration (7). Twenty- Therapeutics Online (http://mct.aacrjournals.org/). two unique FGF family members have been identified, and are Current Address for S.J. Platero: Suso Platero, Head Precision Medicine Oncol- differentially expressed in many, if not all, tissues and organ ogy, Clinical Development Services, Covance Inc., 206 Carnegie Center, Prince- systems, albeit with different patterns and timings (7). FGF ton, NJ 08540-9375. Phone: 609-452-4893; E-mail: [email protected] activity is mediated by four transmembrane receptor tyrosine Corresponding Author: Jayaprakash D. Karkera, Janssen Research and Devel- kinases (FGFRs 14; ref. 8). FGF binding induces FGFR dimer- opment, LLC, 1400 McKean Road, Spring House, PA 19477. Phone: 215-628-5219; ization, leading to phosphorylation of the intracellular tyrosine E-mail: [email protected] kinase domain (9, 10). The dimerization-mediated phosphory- doi: 10.1158/1535-7163.MCT-16-0518 lation leads to the activation of downstream signaling transduc- 2017 American Association for Cancer Research. tion pathways, including FRS2, RAS, RAF, mitogen-activated

www.aacrjournals.org 1717

Karkera et al.

protein kinase (MAPK), PI3K/AKT, signal transducer and activator and FGFR3 chromosomal rearrangements recently identified in of transcription and Phospholipase-C-g (9, 11), which culminate human tumors (26), highlighting the oncogenic potency of these in transcriptional regulation that mediates the functional out- fusions and their sensitivity to FGFR inhibition. Collectively, these comes of FGFR activation. data further inform the oncogenic potency and frequency of FGFR The normal regulation of FGFR signaling is often genetically genetic alterations in human cancers and point to the use of subverted to constitutively activate the pathway in a variety of specific FGFR inhibitors as a rational approach to treating patients malignancies. FGFR activation is attributed to gene amplification, harboring these genetic aberrations. chromosomal translocation, alternative splicing, or point mutations. Dramatic gene amplification of FGFR2 was originally Materials and Methods observed in gastric tumor cell lines, highlighting the oncogenic Experimental gastric tumor models potential of the receptor family (12). Subsequently, FGFR family All animal experiments were performed under approved alterations have been observed in several other malignancies, protocols of the Institutional Animal Use and Care Committee including breast cancer (FGFR1 amplification; ref. 13), bladder of Janssen Pharmaceuticals, Inc. Human gastric cancer xeno- (FGFR3 mutations; refs. 14–16), colorectal (FGFR1 amplification, grafts were established by injection of 106 SNU-16 cells (Amer- FGFR2 and FGFR3 mutations; ref. 17), and hematologic cancers ican Type Culture Collection) from culture into male nude Nu/ (FGFR3 overexpression; refs. 18, 19). More recently, FGFR1 gene Nu rats (nude rate—Crl: NIH-Foxn1(mu) from Charles River amplification has also been observed in 10% of squamous cell Laboratories). Tumors were grown for 18 days, followed by 10 carcinoma (SCC) lung cancer patient samples, but not in days of dosing with JNJ-42883919 (precursor of the clinical lung adenocarcinoma (20). Notably, this aberration has been candidate erdafitinib) or vehicle. The rats were subsequently observed only in tumor samples from patients with a history of sacrificed, and tumors were harvested and prepared into forma- smoking. Approximately 24% of all lung cancers are of the SCC lin-fixed paraffin-embedded (FFPE) blocks. SNU-16 human subtype, which is most common in patients with smoking his- gastric carcinoma cells (106 cells) were fixed in 10% neutral- tories (20–22). Unlike other forms of non–small cell lung cancer buffered formalin and also processed for FFPE blocks. Tumor (NSCLC), for which there are molecularly targeted therapies sections and cell blocks were cut at a thickness of 4 mmand available, SCC treatment options are generally limited to con- placed on silanized glass slides. Antigen retrieval was carried out ventional chemotherapy (23). FGFR1 is one of the first actionable in AR10 citrate buffer (Biogenex) for 20 minutes at 120 Cina genetic targets to be identified as having therapeutic potential in Model 2100 Retriever (Prestige Medical). Tumor sections were SCC lung cancer, as evidenced in preclinical studies of FGFR1 incubated with rabbit monoclonal anti-human pMAPK anti- inhibition showing regression of FGFR1 amplified tumors in a body or rabbit monoclonal anti-human p-S6 antibody (Cell xenograft model (20, 22). FGFR1 may also be a potential ther- Signaling Technology) at room temperature for 45 minutes. Cell apeutic target in other smoking-related malignancies (22, 24). block sections were incubated with rabbit polyclonal anti- FGFR translocations have also been recently observed in a human pFGFR3 (Santa Cruz Biotechnology, Inc.) at room broad range of malignant cell lines, including glioblastoma, temperature for 45 minutes. The Dako EnVision þ DAB rabbit NSCLC, breast cancer, bladder cancer, and cholangiocarcinoma detection system (Dako NorthAmerica,Inc.)wasusedas (6, 25, 26); however, they largely remain functionally unchar- described by the manufacturer, and sections were counter- acterized. FGFR2 was the first family member found to be acti- stained in hematoxylin. All staining was carried out on a Bio- vated by chromosomal rearrangement (FGFR2–FRAG1)ina genex i6000 Automated Staining System. Slides were scanned functional screen to identify oncogenes in osteosarcoma (27). with an Aperio ScanScope (Leica Biosystems), and images were Recently, an FGFR3–TACC3 translocation has been reported in analyzed for antigen content by positive pixel quantitation lung cancer (28, 29). In general, FGFR translocations result in using ImageScope software. constitutive activation of FGFR, due to fusion of the FGFR kinase domain with the N-terminal portion of an unrelated gene harboring domain that promote constitutive receptor dimerization. Fluorescence in situ hybridization (FISH) This dimerization results in receptor phosphorylation and down- The tissue microarrays (TMA) for human NSCLC, prostate, stream signaling leading to oncogenic transformation (6, 25, 27). colorectal, and breast (ER/PR-positive) cancer were constructed The identification of genetic alterations in multiple FGFR and manufactured at ORDIS Biomed using tissues from the family members in human cancers highlights pan-FGFR inhibi- Biobank of the Medical University of Graz, the Lung Biobank of tion as a promising therapeutic approach in a variety of malig- the Institute of Pathology, and the Medical University of nancies. Several small molecule FGFR inhibitors with differing Innsbruck. Tissue blocks and the TMAs for malignant melano- selectivity and potency profiles, such as brivanib, dovitinib, AZD- ma, ovarian, and breast cancer were obtained from Asterand 4547, NVP-BGJ398, and erdafitinib, are in various stages of Bioscience. The TMAs were constructed with triplicate 0.6-mm clinical development. Here we determined the sensitivity of diameter cores extracted from FFPE patient tumor samples, and molecularly annotated panels of cancer cell lines to FGFR pathway tissue cores derived from canis lupus kidney were used as inhibition, assessed the frequency of select FGFR alterations in orientation markers. Collection of tumor specimens was tumor tissue samples from multiple cancers, and confirmed approved by local Institutional Review Boards of Innsbruck pathway inhibition in clinical specimens from patients treated and Graz, Austria. with erdafitinib. These analyses revealed a high frequency of FGFR Dual-color FISH probes for estimation of FGFR gene locus gene amplification in a variety of malignancies and tumor cell amplification were designed, produced, and obtained from lines. Cell lines harboring FGFR alterations were dependent on Kreatech Diagnostics (acquired by Leica Biosystem). The FGFR1 the active pathway for survival and therefore sensitive to treatment probe was specific for locus 8p12 and SE8; the FGFR2 probe with erdafitinib. We also extended this characterization to FGFR2 was specific for locus 10q26 and SE10; and the FGFR3 probe

1718 Mol Cancer Ther; 16(8) August 2017 Molecular Cancer Therapeutics

Predictive and Pharmacodynamic Biomarkers for Erdaﬁtinib

was specific for locus 4p16 and the chromosome-4 centromere 14 days, colonies were fixed and stained with 0.1% cresyl crystal (SE4). For estimation of FGFR4 gene locus amplification, a violet (Sigma-Aldrich). The number of colonies was determined probe specificforNSD1 (5q35) was used as a surrogate for the microscopically by manual counting from triplicate wells for each FGFR4 gene locus (5q33) and an hTERT (5p153) probe was cell line. used as a hybridization control. Probe hybridization was executed per standard procedures with Immunoblotting analysis optimized pretreatment conditions for the individual tissue types. Fusion-overexpressing RK3E cells were plated in complete In brief, 1.5–5-mm sections of TMAs were deparaffinized at 65C growth medium, serum starved overnight, then re-fed with for 30 minutes and washed with Hemo-De solvent. Tissue pre- 0.5% FBS growth media. Cells were treated with 1 mmol/L of fi treatment was executed with 0.2 N hydrochloric acid (20–45 erda tinib and comparator compounds in the presence of minutes, room temperature), followed by microwave heating ligands for 1 hour. For immunoblotting, whole-cell lysates fi (600 W, 20 minutes) and a final protease (Abbott) treatment were collected in RIPA buffer (Thermo Fisher Scienti c), and (15–75 minutes, 37C). For codenaturation, mixed probes (i.e., sample protein concentration was assayed using BCA Protein fi FGFR1/SE8 and FGFR2/SE10) were applied on each section and Assay (Thermo Fisher Scienti c). Equal amounts of protein incubated at 80C for 5 minutes. Subsequent hybridization was (30 mg per lane) were loaded onto 4-12% Bis-Tris gels (Invitro- performed at 38C–42C overnight (16–18 hours). Post-hybrid- gen) followed by SDS-page. Proteins were transferred to nitro- ization washings were performed with 2 saline-sodium citrate/ cellulose membranes and probed with antibodies against p- 0.3% NP-40 detergent (72C, 5 minutes). Finally, slides were FGFR (#3476), total-FGFR2 (#11835), p-MAPK (#4370), total- dehydrated, stained with 40, 6-diamidino-2-phenylindole, and MAPK (#4695), p-S6 (#4858), total S6 (#2317), B-actin (#3700 covered with Vectashield (Vector Laboratories). or #8457) (Cell Signaling Technology), and total-FGFR3 (Santa FISH signals in tumor tissue were enumerated in nonoverlap- Cruz Biotechnology). The membranes were blocked with Odys- sey blocking buffer (Li-Cor Biosciences) for 1 hour at room ping nuclei with complete and intact contours. Target signals (red; i.e., FGFR1 and FGFR2) and control signals (green; i.e., SE8 and temperature, and then incubated overnight at 4 Cinaprimary SE10) were recorded for each individual core. Digital FISH images antibody solution (1:1,000) diluted in Odyssey blocking buff- were captured using a Zeiss Axiophot microscope with an Axio- er. After three washes in tris-buffered saline plus 0.1% Tween 20 cam MRm camera (Zeiss) at a magnification of 1,000. Images (TBST), the membranes were probed with goat anti-mouse or were processed using Adobe Photoshop 7.0 and exported at a donkey anti-rabbit IRDye 680RD or 800CW (Li-Cor Bios- resolution of 150 pixels/inch. ciences) labeled secondary antisera (1: 10,000) in Odyssey Gene amplification was estimated by the gene ratio (number blocking buffer for 1 hour at room temperature. Washes were of target signals vs. control). A gene ratio 2wasdefined as repeated after secondary labeling and the membranes were amplification. When gene locus amplification was restricted to imaged using an Odyssey scanner and the Odyssey 3.0 analyt- fi <20%–50% of the tumor, an average gene ratio of <2was ical software (Li-Cor Biosciences). Effects of erda tinib were computed (i.e., down to 1.4). These cases were classified as compared with AZD4547 and NVP-BGJ398. amplified but presenting with substantial intratumor hetero- Drug response testing for FGFR fusion-overexpressing cell lines geneity. The gene dose of either target or control was computed Fusion-overexpressing RK3E cells were seeded into Cultur- by the ratio of number of target signals versus nuclei evaluated. Plate-96 (Perkin Elmer) well plates (1,000 cells/well) in trip- Agenedose 3wasdefinedasgenedoseelevation(i.e.,FGFR1, licate in complete growth medium plus the ligands FGF-1 and FGFR2, FGFR3,orFGFR4) or polysomy of the probed chro- FGF-2 (R&D Systems Inc.). After 24 hours, cells were serum mosome (i.e., SE4, SE8, or SE10). Student's t-test was used to starved overnight then re-fed with 0.5% FBS growth media plus determine statistical differences between FGFR genetic profiles FGF-1 and FGF-2. Seventy-two hours after plating, cells in different types of prostate cancer and between SCC and were treated with an 18-point 1:3 dilution series, starting at adenocarcinoma in NSCLC. 10 mmol/L of erdafitinib, AZD4547, and NVP-BGJ398. The Microtiter plates were then incubated for 72 hours and assayed FGFR fusion overexpressing cell lines for adenosine triphosphate content using the Cell Titer-Glo RK3E cells (American Type Culture Collection), from rat kidney Luminescent Cell Viability assay (Promega Corp.) following epithelial cells, were cultured in DMEM supplemented with 10% the manufacturer's instructions with modifications. Briefly, the FBS and 1% antibiotics (Invitrogen). FGFR fusion gene constructs cells were allowed to equilibrate to room temperature, at which were designed and cloned into the pReceiver expression vector time a 1:1 mixture of Cell Titer-Glo reagent was added. The cells (Genecopoeia). Clones were transfected into RK3E cells using the were then placed on an orbital shaker for 2 minutes and Amaxa Cell Line Nucleofector (Lonza) following the manufac- incubated for 10 minutes at room temperature to stabilize the turer's protocol. The stably transfected cells were selected in luminescent signal. The luminescence was quantified and the complete medium with 800 mg/mL of G418 (Invitrogen). Over- measurements were then conducted using an Envision Multi- expression of the fusions in the stably transfected cells was label plate reader (Perkin Elmer). IC values were calculated fi 50 con rmed by real-time PCR and immunoblotting. using GraphPad Prism 5.0 (GraphPad Software Inc.).

Colony formation assay Test and authentication of cell lines Anchorage-independent growth of the fusion gene-overexpres- The RK3E and SNU-16 cell lines were purchased directly sing cells was tested. One mL culture medium with 0.8% low from American Type Culture Collection on September 4, 2014. melting point agarose was ﬁrst plated into each of three wells of a The cell lines were tested for Mycoplasma using the MycoAlert kit six-well plate. After the agar solidiﬁed, each well received another (Lonza) before they were used in any experiments. After four 1 mL of 0.4% agar in culture medium containing 100 cells. After passages, the cell lines were aliquoted and frozen in vials as

www.aacrjournals.org Mol Cancer Ther; 16(8) August 2017 1719

Karkera et al.

stocks. The transformed stable cell lines were last tested in July rabbit mAb #4370 (R647) from Cell Signaling Technology was 2015. No independent cell line authentication was performed used as the detecting antibody. on these cell lines.

Phase I clinical trial Results Erdafitinib, or JNJ-42756493 (Supplementary Fig. S1), dis- Predictive and pharmacodynamic biomarkers covered in collaboration with Astex Pharmaceuticals, is an SNU-16 cells, which carry a high-level FGFR2 amplification, investigational agent currently in clinical development. The were sensitive to the FGFR inhibitor JNJ-42883919, and FGFR dose escalation and expansion phase I study has been previ- phosphoprotein levels decreased upon treatment with the study ously described (30). drug, as measured by immunohistochemical staining of FFPE tumor cells (Fig. 1A and B). Quantification of the pan-FGFR signal Immunohistochemical analysis of tumor in SNU-16 cells treated with the vehicle versus with the FGFR pharmacodynamic biomarkers inhibitor showed a shift from strong labeling to more moderate Tumor biopsies for pharmacodynamic analyses were option- labeling (Fig. 1B). In athymic rats carrying SNU-16 gastric tumor al in the dose escalation phase but mandatory in the dose xenografts, phosphorylated S6 and MAPK (pS6 and pMAPK) confirmation phase. Tumor biopsy samples were collected at levels markedly decreased upon treatment with JNJ-42883919 pretreatment (within 28 days prior to the firstdoseofstudy (Fig. 1C–F), thus confirming downstream signal transduction drug) and posttreatment (day 1 of cycle 2 7 days). Immu- endpoint modulation that could be explored as pharmacody- nohistochemical analysis was performed on the Ventana namics marker of response in a clinical trial. Benchmark platform using a standard protocol. The phos- To determine the frequency of FGFR amplifications in pho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (D13.14.4E) human cancers, tumor microarrays were analyzed by FISH for

Figure 1. Pharmacodynamic markers of FGFR inhibition. SNU-16 cells stained with anti–pan-FGFR. A, Vehicle; B, after treatment with JNJ-42883919. Tumors harvested from rats carrying SNU-16 xenografts were stained with anti-pS6 (C and D) and anti-pMAPK (E and F). Tumors of vehicle control-treated animals are shown in C and E; xenograft tumor of animals treated with JNJ- 42883919 is shown in D and F.

1720 Mol Cancer Ther; 16(8) August 2017 Molecular Cancer Therapeutics

Predictive and Pharmacodynamic Biomarkers for Erdaﬁtinib

Figure 2. Fluorescence in situ hybridization analysis of FGFR gene amplifications. A, FGFR1; B, FGFR2 gene locus amplification in NSCLC; C, high-grade FGFR1 gene locus amplification in ER/PR- positive breast cancer; D, high-grade FGFR2 gene locus amplification in triple-negative breast cancer; E, FGFR1; and F, FGFR2 low- to moderate-grade amplification in prostate cancer; G, FGFR4 low-grade level gene amplification in breast cancer.

thepresenceandabundanceofthe four FGFR genes (Fig. 2). In detected in either colon cancer or malignant melanoma tumor breast tumor samples FGFR1, FGFR2, FGFR3,andFGFR4 were samples (Supplementary Table S1). amplified at a frequency of 11.7% (29/248), 6.6% (13/198), 3% (1/29), and 8.6% (15/175), respectively. In NSCLC tumor Characterization of FGFR fusion genes samples FGFR1, FGFR2, FGFR3,andFGFR4 were amplified at a One tumor cell line, RT-4, in the selected tumor cell line panel frequency of 21.6% (45/208), 5.6% (11/195), 5.3% (1/19), analyzed carries an FGFR3 C-terminal translocation and was and 5.1% (5/99), respectively (Supplementary Table S1). In sensitive to both JNJ-42541707 (IC50 ¼ 27 nmol/L) and erdafi- prostate tumor samples only FGFR1 and FGFR2 amplifications tinib (IC50 ¼ 0.19 nmol/L; Supplementary Table S2). Based on were measured and the frequencies were 14.6% (14/96) and this finding, we assessed the capability of other clinically relevant 1.1% (1/94), respectively. While in ovarian cancer FGFR1 and FGFR translocations to transform normal RK3E cells and confer FGFR2 were amplified at a frequency of 2.9% (1/35) and 5.6% sensitivity to erdafitinib (26). FGFR3:TACC3v1, FGFR3:TACC3v3, (2/36), respectively. No FGFR family gene amplifications were FGFR3:BAIP2L1, FGFR2:BICC1, FGFR2:AFF3, FGFR2:CASP7,

www.aacrjournals.org Mol Cancer Ther; 16(8) August 2017 1721

Karkera et al.

Figure 3. Characterization of FGFR2 and FGFR3 fusion gene transformation activity and sensitivity to FGFR inhibitors. A, Effect of expression of FGFR2 and FGFR3 fusion constructs in RK3E cells in an anchorage-independent colony assay. B, Effect of FGFR inhibitors on the MAPK signaling pathway in RK3E cells expressing stable FGFR2 and FGFR3 fusion constructs and empty vector.

FGFR2:CCDC6, and FGFR2:OFD1 translocations were cloned FGFR inhibitors tested reduced the levels of MAPK and S6 phos- into vectors and transfected into untransformed RK3E recipient phorylation induced by the transfection of RK3E cells with FGFR3 cells to determine the oncogenic potential of the FGFR alterations and FGFR2 fusion proteins (Fig. 3B). In contrast, the effect of in an anchorage-independent growth assay. All FGFR-activating AZD4547 and NVP-BGJ398 on proliferation of these FGFR fusion translocations tested conferred the ability to promote soft-agar gene-expressing cells was weaker than the growth inhibition by growth compared with wild-type FGFRs (Fig. 3A). The transfor- erdaﬁtinib (Supplementary Table S2). The detection of total mation activity of these translocations was associated with con- FGFR2 by the FGFR2 antibody was not successful in any of the stitutive activation of FGFR and downstream signaling. The effects FGFR2 fusion harboring RK3E cells, but we could detect the HA of three pan-FGFR inhibitors, erdaﬁtinib, AZD4547, and NVP- tag expression in each of the FGFR2-overexpressing cell lines, thus BGJ398, on the MAPK signaling pathway in RK3E cells expressing suggesting that the FGFR2 protein was expressed (Supplementary these FGFR fusion genes was assessed by immunoblotting. All Fig. S2). The failure to detect FGFR2 by FGFR2 antibody could

1722 Mol Cancer Ther; 16(8) August 2017 Molecular Cancer Therapeutics

Predictive and Pharmacodynamic Biomarkers for Erdaﬁtinib

potentially be due to the masking of the specific epitope the who had failed three prior therapies was treated with erdafitinib. FGFR2 antibody targets caused by the tertiary conformation of the This patient had tumor shrinkage of 38% (Fig. 5) with a confirmed fusion protein. Overall, these results demonstrate that the FGFR2 partial response by RECIST criteria (33) and stayed on treatment and FGFR3 translocations tested are oncogenic and confer sensi- (at 9 mg daily) for about 10 months. The tumor shrinkage was tivity to FGFR inhibitors. calculated based on target lesion measurements at baseline and posttreatment efficacy assessment time point. The FGFR3–TACC3 Assessment of response markers in patients treated translocation was detected in this patient's tumor sample, extend- with erdafitinib ing the observation that these alterations are a predictor of To confirm the utility of pharmacodynamic markers identified sensitivity to erdafitinib clinically. in this study, immunohistochemical analysis for phospho-Erk (pErk), a downstream mediator of the FGFR pathway, was carried out on pretreatment and posttreatment biopsies from patients Discussion treated with erdafitinib (30). Samples were obtained from two A detailed analysis of sensitive and resistant cell lines identified patients enrolled in the phase I clinical trial of erdafitinib. One FGFR gene amplification and translocations as potential markers breast cancer patient with archival tumor harboring an FGFR1 of response to FGFR inhibitors. FGFR2 and FGFR3 translocations amplification was treated with erdafitinib at 12 mg daily for 2 were shown to be potent oncogenes and resulted in constitutive weeks and biopsied after 7 days of drug interruption (Fig. 4A). An FGFR phosphorylation and downstream signaling. Transformed approximate 40% reduction in pErk posttreatment was observed. cells expressing these FGFR alterations were sensitive to FGFR In another patient with pleural mesothelioma of unknown FGFR small molecule inhibition with several inhibitors in clinical status (Fig. 4B), a 45% reduction in pErk was observed after 3 development, but erdafitinib showed the highest potency in a weeks of treatment with erdafitinib at 6 mg daily. Other phar- comparative analysis. Interestingly, a patient whose tumor har- macodynamic makers such as calcium, phosphate, soluble bored an FGFR translocation, FGFR3–TACC3, showed a durable FGF23, soluble VEGFR, soluble FGFR2, 3 and 4, and vitamin partial response to erdafitinib by RECIST criteria, demonstrating D, were also measured from sera of patients undergoing the trial. that the effects observed in cell lines were also applicable to cancer Of those, only phosphate levels showed significant dose propor- patients in the clinic. tional changes upon treatment (30). The increase of serum FGFR amplifications were shown to be frequent genetic events phosphate concentrations in response to erdafitinib, due to renal in a variety of malignancies. Preclinically, sensitivity was also seen tubular FGFR inhibition (31), extends similar findings on serum with FGFR amplifications in tumor cell lines and xenograft phosphate levels observed for other FGFR inhibitors (32). models, where a decrease in signaling downstream of FGFR was In the phase I study (30), a 52-year-old bladder cancer patient, demonstrated upon administration of FGFR inhibitors. A recent enrolled in the dose escalation cohort, with metastases to the lung report of positive clinical activity of the FGFR inhibitor AZD4547

Figure 4. Pharmacodynamic biomarkers modulation in patients receiving erdafitinib. Phospho-Erk staining in A, a patient with breast cancer receiving erdafitinib, 12 mg daily; B, a patient with pleural mesothelioma receiving erdafitinib, 6 mg daily. Left and right panels represent pre-dose and post-dose conditions, respectively.

www.aacrjournals.org Mol Cancer Ther; 16(8) August 2017 1723

Karkera et al.

A B

C D

Figure 5. CT scan images of a patient with FGFR3–TACC3-translocated bladder cancer with a conﬁrmed partial response in lung metastases (marked with white arrows on baseline scans) responding to erdaﬁtinib. A and C are baseline lung images, and B and D are posttreatment lung images at week 6 for the same patient.

in gastric and breast cancer patients with high-level amplification clinical models, downregulation of the pharmacodynamic bio- of FGFR2 is very encouraging (34). Interestingly, in the current markers (pErk and pS6) was demonstrated in the SNU-16 study the high-level amplification of FGFR2 was observed in model after treatment with an FGFR inhibitor. These endpoints breast, NSCLC, and ovarian cancers at a frequency of 6.6%, were subsequently shown to be modulated in two patients 5.6%, and 6.0%, respectively. But, in a recent report (35) using receiving erdafitinib, confirming the activity of the inhibitor on next-generation sequencing method, researchers reported the suppressing FGFR-mediated signal transduction (36). Serum amplification of FGFR2 in breast, NSCLC, and ovarian cancers phosphate homeostasis has been reported to be controlled by at a frequency of 0.8%, 1.7%, and 0.3%, respectively. Similarly, a FGFR activity in the renal tubules through the FGF23–klotho high-level amplification of FGFR1 was reported in breast, ovarian, signaling axis and that genetic deletion of FGF23 results in and NSCLC at a frequency of 14%, 5%, and 9% (NSCLC squa- upregulation of serum phosphate (32). We showed that erda- mous) or 4% (NSCLC adenocarcinoma; ref. 35), respectively, fitinib regulating serum phosphate levels in a dose dependent whereas our data using the FISH method showed a frequency of manner in patients was managed clinically through chelation 11.7%, 2.9%, and 21.6%, respectively. The discrepancy of the therapy and was not associated with significant adverse events findings may be due to the different methodologies used in the (30), an observation consistent with this being an on-mecha- assessment of the FGFR high-level amplifications. The SNU-16 nism class effect (26). Interestingly, FGFR-mediated phosphate gastric cell line with high-level of FGFR2 amplification has been increases provide an interesting on-mechanism endpoint for shown to be sensitive to erdafitinib, which is in concordance with monitoring clinical target engagement to support subsequent the data reported for AZD4547 (34). However, overexpression of dose selection. FGFRs in transformation assays is not sufficient to confer a potent In conclusion, the data from this study credential FGFR ampli- oncogenic effect without ligand expression (27), suggesting that fications and C-terminal translocations as dominant oncogenic FGFR gene amplifications may be less dominant driver pathway events that confer sensitivity to small molecule FGFR inhibitors. events compared with translocations. These FGFR activating alterations resulted in constitutive pathway A key aspect of drug development is generating clinical data activation that was observed preclinically and clinically. On-target to support what dose levels are taken forward into late clinical FGFR-mediated changes in phosphate regulation were observed development. To this end, we examined FGFR-mediated signal clinically and a patient expressing a FGFR3–TACC3 translocation transduction and functional endpoints to determine maximal received clinical benefit from treatment with erdafitinib. Col- target engagement as a component of dose selection. In pre- lectively, these observations highlight erdafitinib as a selective

1724 Mol Cancer Ther; 16(8) August 2017 Molecular Cancer Therapeutics

Predictive and Pharmacodynamic Biomarkers for Erdaﬁtinib

FGFR family inhibitor and support its ongoing clinical develop- Writing, review, and/or revision of the manuscript: J.D. Karkera, G.M. Car- ment in patients with activating FGFR genetic alterations. dona, K. Bell, D. Gaffney, J.C. Portale, A. Santiago-Walker, P. King, M. Sharp, R. Bahleda, F.R. Luo, J.D. Alvarez, M.V. Lorenzi, S.J. Platero Administrative, technical, or material support (i.e., reporting or organizing Disclosure of Potential Conflicts of Interest data, constructing databases): J.D. Karkera, G.M. Cardona, K. Bell, D. Gaffney, F.R. Luo is an employee of Janssen and has an ownership interest in the J.C. Portale, M. Sharp, J.D. Alvarez company stock. J.D. Alvarez is a senior scientific director at Janssen Study supervision: J.D. Karkera, F.R. Luo, M.V. Lorenzi, S.J. Platero Research and Development and has ownership interest (including patents) in the same. J.D. Karkera and A. Santiago-Walker have an ownership Acknowledgments interest in the company stock (including patents). No potential conflicts This study was funded and supported by Janssen Research and Develop- of interest were disclosed by the other authors. ment, LLC. The authors would like to thank Harry Ma for writing and editorial assistance; and Clifford O. Motley, Dan Rhodes, Timothy Perera, Authors' Contributions Chris Takimoto, Vijay Peddareddigari, and Marcus Otte for their support and guidance. Conception and design: J.D. Karkera, G.M. Cardona, K. Bell, D. Gaffney, P. King, R. Bahleda, M.V. Lorenzi, S.J. Platero Development of methodology: J.D. Karkera, G.M. Cardona, K. Bell, D. Gaffney, Grant Support fi P. King, R. Bahleda, J.D. Alvarez, S.J. Platero This study was nancially supported by Janssen Research and Development, Acquisition of data (provided animals, acquired and managed patients, LLC. provided facilities, etc.): J.D. Karkera, G.M. Cardona, K. Bell, D. Gaffney, The costs of publication of this article were defrayed in part by the payment of advertisement J.C. Portale, P. King, R. Bahleda, F.R. Luo, S.J. Platero page charges. This article must therefore be hereby marked in Analysis and interpretation of data (e.g., statistical analysis, biostatistics, accordance with 18 U.S.C. Section 1734 solely to indicate this fact. computational analysis): J.D. Karkera, G.M. Cardona, K. Bell, D. Gaffney, J.C. Portale, C.H. Moy, P. King, M. Sharp, R. Bahleda, F.R. Luo, J.D. Alvarez, Received August 3, 2016; revised January 6, 2017; accepted April 5, 2017; M.V. Lorenzi, S.J. Platero published OnlineFirst April 17, 2017.

References 1. Burstein HJ, Harris LN, Marcom PK, Lambert-Falls R, Havlin K, Overmoyer 15. Kiemeney LA, Sulem P, Besenbacher S, Vermeulen SH, Sigurds- B, et al. Trastuzumab and vinorelbine as first-line therapy for HER2-over- son A, Thorleifsson G, et al. A sequence variant at 4p16.3 expressing metastatic breast cancer: multicenter phase II trial with clinical confers susceptibility to urinary bladder cancer. Nat Genet 2010; outcomes, analysis of serum tumor markers as predictive factors, and 42:415–9. cardiac surveillance algorithm. J Clin Oncol 2003;21:2889–95. 16. van Rhijn BW, van der Kwast TH, Liu L, Fleshner NE, Bostrom PJ, Vis AN, 2. Amado RG, Wolf M, Peeters M, Van Cutsem E, Siena S, Freeman DJ, et al. et al. The FGFR3 mutation is related to favorable pT1 bladder cancer. J Urol Wild-type KRAS is required for panitumumab efficacy in patients with 2012;187:310–4. metastatic colorectal cancer. J Clin Oncol 2008;26:1626–34. 17. Jang JH, Shin KH, Park JG. Mutations in fibroblast growth factor receptor 2 3. Karapetis CS, Khambata-Ford S, Jonker DJ, O'Callaghan CJ, Tu D, Tebbutt and fibroblast growth factor receptor 3 genes associated with human gastric NC, et al. K-ras mutations and benefit from cetuximab in advanced and colorectal cancers. Cancer Res 2001;61:3541–3. colorectal cancer. N Engl J Med 2008;359:1757–65. 18. Chang H, Stewart AK, Qi XY, Li ZH, Yi QL, Trudel S. Immunohistochem- 4. Pao W, Miller V, Zakowski M, Doherty J, Politi K, Sarkaria I, et al. EGF istry accurately predicts FGFR3 aberrant expression and t(4;14) in multiple receptor gene mutations are common in lung cancers from "never smokers" myeloma. Blood 2005;106:353–5. and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc 19. Dvorak P, Dvorakova D, Doubek M, Faitova J, Pacholikova J, Hampl A, Natl Acad Sci U S A 2004;101:13306–11. et al. Increased expression of fibroblast growth factor receptor 3 in CD34þ 5. Grande E, Bolos MV, Arriola E. Targeting oncogenic ALK: a promising BCR-ABLþ cells from patients with chronic myeloid leukemia. Leukemia strategy for cancer treatment. Mol Cancer Ther 2011;10:569–79. 2003;17:2418–25. 6. Williams SV, Hurst CD, Knowles MA. Oncogenic FGFR3 gene fusions in 20. Weiss J, Sos ML, Seidel D, Peifer M, Zander T, Heuckmann JM, et al. bladder cancer. Hum Mol Genet 2013;22:795–803. Frequent and focal FGFR1 amplification associates with therapeutically 7. Ornitz DM, Itoh N. Fibroblast growth factors. Genome Biol 2001;2: tractable FGFR1 dependency in squamous cell lung cancer. Sci Transl Med REVIEWS3005. 2010;2:62ra93. 8. Turner N, Grose R. Fibroblast growth factor signalling: from development 21. Garber ME, Troyanskaya OG, Schluens K, Petersen S, Thaesler Z, Pacyna- to cancer. Nat Rev Cancer 2010;10:116–29. Gengelbach M, et al. Diversity of gene expression in adenocarcinoma of the 9. Mohammadi M, Dikic I, Sorokin A, Burgess WH, Jaye M, Schlessinger J. lung. Proc Natl Acad Sci U S A 2001;98:13784–9. Identification of six novel autophosphorylation sites on fibroblast growth 22. Turner NC, Seckl MJ. A therapeutic target for smoking-associated lung factor receptor 1 and elucidation of their importance in receptor activation cancer. Sci Transl Med 2010;2:62ps56. and signal transduction. Mol Cell Biol 1996;16:977–89. 23. Network NCC. NCCN Clinical Practice Guidelines in Oncology: Non- 10. Plotnikov AN, Schlessinger J, Hubbard SR, Mohammadi M. Structural basis Small Cell Lung Cancer. Version 52015. for FGF receptor dimerization and activation. Cell 1999;98:641–50. 24. Freier K, Schwaenen C, Sticht C, Flechtenmacher C, Muhling J, Hofele 11. Hart KC, Robertson SC, Kanemitsu MY, Meyer AN, Tynan JA, Donoghue C, et al. Recurrent FGFR1 amplification and high FGFR1 protein DJ. Transformation and Stat activation by derivatives of FGFR1, FGFR3, expression in oral squamous cell carcinoma (OSCC). Oral Oncol and FGFR4. Oncogene 2000;19:3309–20. 2007;43:60–6. 12. Kunii K, Davis L, Gorenstein J, Hatch H, Yashiro M, Di Bacco A, et al. 25. Singh D, Chan JM, Zoppoli P, Niola F, Sullivan R, Castano A, et al. FGFR2-amplified gastric cancer cell lines require FGFR2 and Erbb3 signal- Transforming fusions of FGFR and TACC genes in human glioblastoma. ing for growth and survival. Cancer Res 2008;68:2340–8. Science 2012;337:1231–5. 13. Elbauomy Elsheikh S, Green AR, Lambros MB, Turner NC, Grainge MJ, 26. Wu YM, Su F, Kalyana-Sundaram S, Khazanov N, Ateeq B, Cao X, et al. Powe D, et al. FGFR1 amplification in breast carcinomas: a chromogenic in Identification of targetable FGFR gene fusions in diverse cancers. Cancer situ hybridisation analysis. Breast Cancer Res 2007;9:R23. Discov 2013;3:636–47. 14. Dodurga Y, Tataroglu C, Kesen Z, Satiroglu-Tufan NL. Incidence of 27. Lorenzi MV, Horii Y, Yamanaka R, Sakaguchi K, Miki T. FRAG1, a gene that fibroblast growth factor receptor 3 gene (FGFR3) A248C, S249C, potently activates fibroblast growth factor receptor by C-terminal fusion G372C, and T375C mutations in bladder cancer. Genet Mol Res through chromosomal rearrangement. Proc Natl Acad Sci U S A 2011;10:86–95. 1996;93:8956–61.

www.aacrjournals.org Mol Cancer Ther; 16(8) August 2017 1725

Karkera et al.

28. Majewski IJ, Mittempergher L, Davidson NM, Bosma A, Willems SM, signaling ameliorates FGF23-mediated hypophosphatemic rickets. J Bone Horlings HM, et al. Identification of recurrent FGFR3 fusion genes in Miner Res 2013;28:899–911. lung cancer through kinome-centred RNA sequencing. J Pathol 2013; 33. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. 230:270–6. New response evaluation criteria in solid tumours: revised RECIST guide- 29. Wang R, Wang L, Li Y, Hu H, Shen L, Shen X, et al. FGFR1/3 tyrosine kinase line (version 1.1). Eur J Cancer 2009;45:228–47. fusions define a unique molecular subtype of non-small cell lung cancer. 34. Pearson A, Smyth E, Babina IS, Herrera-Abreu MT, Tarazona N, Clin Cancer Res 2014;20:4107–14. Peckitt C, et al. High-level clonal FGFR amplification and response 30. Tabernero J, Bahleda R, Dienstmann R, Infante JR, Mita A, Italiano A, et al. to FGFR inhibition in a translational clinical trial. Cancer Discov Phase I dose-escalation study of JNJ-42756493, an oral pan-fibroblast 2016;6:838–51. growth factor receptor inhibitor, in patients with advanced solid tumors. 35. Helsten T, Elkin S, Arthur E, Tomson BN, Carter J, Kurzrock R. The FGFR J Clin Oncol 2015;33:3401–8. landscape in cancer: analysis of 4,853 tumors by next-generation sequenc- 31. YanochkoGM,VitskyA,HeyenJR,HirakawaB,LamJL,MayJ,etal.Pan- ing. Clin Cancer Res 2016;22:259–67. FGFR inhibition leads to blockade of FGF23 signaling, soft tissue 36. Fischer H, Taylor N, Allerstorfer S, Grusch M, Sonvilla G, Holzmann K, et al. mineralization, and cardiovascular dysfunction. Toxicol Sci 2013;135: Fibroblast growth factor receptor-mediated signals contribute to the malig- 451–64. nant phenotype of non-small cell lung cancer cells: therapeutic implica- 32. Wohrle S, Henninger C, Bonny O, Thuery A, Beluch N, Hynes NE, et al. tions and synergism with epidermal growth factor receptor inhibition. Mol Pharmacological inhibition of fibroblast growth factor (FGF) receptor Cancer Ther 2008;7:3408–19.

1726 Mol Cancer Ther; 16(8) August 2017 Molecular Cancer Therapeutics

Oncogenic Characterization and Pharmacologic Sensitivity of Activating Fibroblast Growth Factor Receptor (FGFR) Genetic Alterations to the Selective FGFR Inhibitor Erdafitinib

Jayaprakash D. Karkera, Gabriela Martinez Cardona, Katherine Bell, et al.

Mol Cancer Ther 2017;16:1717-1726. Published OnlineFirst April 17, 2017.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-16-0518

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2017/04/15/1535-7163.MCT-16-0518.DC1

Cited articles This article cites 35 articles, 19 of which you can access for free at: http://mct.aacrjournals.org/content/16/8/1717.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/16/8/1717.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/16/8/1717. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.