1080 | DO NOT POST Altered response to BET Bromodomain inhibitors JQ1 and I-BET-762 targeting c-Myc in erdafitinib-resistant endometrial carcinoma cell line AN3 CA

Wilhelmina E. van Riel, Janneke J. T. M. Melis, Demi H. J. Vogels, Melanie Thielens, Winfried R. Mulder, Jeffrey J. Kooijman, Rogier C. Buijsman, Guido J. R. Zaman. Netherlands Translational Research Center B.V. (NTRC), Kloosterstraat 9, 5349 AB Oss, The Netherlands | T: +31 412 700 500 E: [email protected] W: www.ntrctx.com - www.oncolines.com

Results Table 2. Genes reported in FGFR inhibitor resistance are not the cause of resistance to erdafitinib in Introduction the AN3 CA resistant cell lines. Inhibition of indicated targets by small molecule inhibitors revealed unaltered A JQ1 B I-BET-762 response (blue) or decreased sensitivity (green) of the resistant lines compared to the parental cell line. IC50 values • FGFR alterations are common in multiple cancer types, including breast were calculated from dose-response curves. ParentalParental ParentalParental 120 120 120 120Parental Parental cancer, non-small cell lung cancer and endometrial cancer. A erdafitinib B C 120 AN3 CA_resist_02AN3 CA_resist_02120 AN3 CA_resist_02AN3 CA_resist_02 IC50 values (nM) 100 100 100 100AN3 CA_resist_02 AN3 CA_resist_02 Compound name Target(s) 100 AN3 CA_resist_05AN3 CA_resist_05100 AN3 CA_resist_05AN3 CA_resist_05 120 AN3 CA_resist_05 AN3 CA_resist_05 Parental Parental Parental Resist_02 Resist_05 Resist_04 80 80 80 80 • The pan-FGFR inhibitor erdafitinib is used to treat advanced or metastatic 120 120 80 AN3 CA_resist_04AN3 CA_resist_0480 AN3 CA_resist_04AN3 CA_resist_04 AN3 CA_resist_04 AN3 CA_resist_04 100 AN3 CA_resist_02 AN3 CA_resist_02 60 60 60 60 100 100 erdafitinib pan-FGFR (1-4) 16 11433 10789 11595 60 60 80 AN3 CA_resist_05 AN3 CA_resist_05 40 40 40 40 urothelial carcinoma harboring genetic FGFR2 or FGFR3 alterations. 80 80 40 40 60 AN3 CA_resist_04 AN3 CA_resist_04 infigratinib FGFR1/2/3 84 5520 4663 6640 20 20 20 20 20 20 60 60 viability, % of untreated viability, % of untreated viability, % of untreated viability, % of untreated

0 0 viability, % of untreated 0 0 viability, % of untreated • After a good initial response, patients rapidly develop clinical resistance to 40 0 0 40 40 pemigatinib FGFR1/2/3 57 2352 3379 4608 -9-9 -8 -8 -7 -7 -6 -6-5-5 -4 -4 -9-9 -8 -8 -7 -7 -6 -6-5-5 -4 -4 -9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 20 concentration,concentration, log M log M concentration,concentration, log M log M 20 20 concentration, log M concentration, log M viability, % of untreated viability, % of untreated viability, % of untreated derazantinib FGFR1/2/3 1722 5780 5812 4946 erdafitinib and have a progression-free survival of only a few months. 0 0 0 -9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 AZD4547 FGFR1/2/3 97 14605 16886 13119 • Insight into acquired resistance mechanisms to erdafitinib treatment aids concentration, log M concentration, log M concentration, log M FGFR, VEGFR1/2/3, Figure 5. In vitro drug sensitivity analysis confirms that the resistant cell lines have increased 155 30966 25903 18450 in the development of new therapies. PDGFR,c-Kit, c-Fms sensitivity to c-Myc targeting. BET Bromodomain inhibitors JQ1 and I-BET-762 were used to FGFR1, Abl, PDGFRa, indirectly target c-Myc. Dose-response curves of the AN3 CA parental and resistant lines exposed to (A) D derazantinib E AZD4547 24 800 937 1120 • Known resistance mechanisms to FGFR inhibitors include acquired Derazantinib AZD4547 VEGFR2, Src JQ1 and (B) I-BET-762. VEGFR1/2/3, Parental Parental 237 3529 3806 ND 120 120 Erdafitinib FGFR1/2/3, PDGFRa/ß mutations in FGFRs, alternative signaling via other receptors, i.e., MET, AN3 CA_resist_02 AN3 CA_resist_02 100 100 120 PDGF-Rs, VEGFRs, AN3 CA_resist_05 AN3 CA_resist_05Parental 80 80 ERBB2/3 or EGFR, activation of internal signaling pathways and epithelial- AN3 CA_resist_04 100 AN3 CA_resist_04AN3 CA_resist_02 c-KIT, RET, CD114, 939 4080 5017 3442 60 60 CD135 80 AN3 CA_resist_05 A B 40 40 mesenchymal transition. 60 AN3 CA_resist_04 EGFR 2849 2891 2929 3014 C-Myc 20 20 1.5 viability, % of untreated viability, % of untreated 40 AN3 CA • We developed erdafitinib-resistant cell linesin vitro and investigated the 0 0 EGFR, ERBB2 14201 15803 11989 14990 -9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 20 1.0 Parental Resist_02 Resist_05 Resist_04

concentration, log M concentration,viability, % of untreated log M EGFR, ERBB2 >3160 >3160 2800 >3160 mechanism of acquired resistance to the drug. 0 2 6 18 24 0 2 6 18 24 0 2 6 18 24 0 2 6 18 24 hours erdafitinib 0.5 Figure 2. Cell lines selected for resistance against erdafitinib are cross-resistant to other ERBB2 24686 16384 20529 29794 (fold change) c-Myc

FGFR inhibitors. Dose-response curves of the AN3 CA parental and resistant lines exposed to pan- Relative mRNA expression 0.0 MET, ALK 1903 2039 1782 2472 l 2 5 4 2 5 4 Histone H3 a 0 0 0 0 0 0 t t t t t t AN3 CA ent s s s s s s ar si si si si si si FGFR1-4 inhibitor erdafitinib, FGFR1/2/3 inhibitors infigratinib (B), pemigatinib (C), derazantinib (D) and VEGFR2, c-Met, Ret, p re re re re re re 3088 >31600 >31600 19570 - erdafitinib + erdafitinib Experimental approach AZD4547 (E). The IC50 values are listed in Table 2. Kit, Flt-1/3/4, Tie2, AXL MEK 50 6631 23751 286 • Endometrial adenocarcinoma cell line AN3 CA, harboring FGFR2 gain- PD 0325901 MEK 1146 1457 2619 710 Figure 6. Alterations in c-Myc stability are seen in the resistant cell lines. (A) No alterations in AMPK activator c-Myc expression are observed in mRNA expression as determined by qPCR analysis. Expression levels of-function mutation N549K, was cultured with gradually increasing metformin >31600 >31600 >31600 >31600 were normalized for expression of GAPDH, beta-actin, HPRT1 and RPS18, averaged and related to the A MET B AN3 CA C (mTORC inhibition) crizotinibCrizotinib KRAS-SOS1 interac- parental line. (B) Immunoblot showing a time course of erdafitinib treatment on parental and resistant concentrations of erdafitinib. 600000 Bay-293 2654 2485 2294 2738 Parental 300000 120 tion inhibitor cell lines reveals altered stability of c-Myc in resistant cell lines.

Hs 746TSNU-5Parental AN3 CA_resist_02 • The concentration was increased when cells were recovered from the 600 Resist_02Resist_05Resist_04 100 ipatasertib AKT 114 480 634 780 AN3 CA_resist_05 Crizotinib 80 400 MET (low exposure) AN3 CA_resist_04MK-2206 AKT 540 8390 6607 4693 60 Parental

previous erdafitinib concentration and were in exponential growth phase. (fold change) 120 200 Resist_02 100 40 Resist_05 dactolisib PI3K/mTOR 51 112 102 145 80 • The resistant cell lines were compared to the parental cell line on Relative mRNA expression 0 20 Resist_04 Conclusions 5 l 2 5 4 2 5 4 60

T viability, % of untreated 6 - a 0 0 0 0 0 0 MET (high exposure) U t t t t t t N ent s s s s s s 0 mTOR 59 138 53 208 s 74 S ar si si si si si si 40 H p re re re re re re -9 -8 -7 -6 -5 -4 20

expression of targets, response to small molecule inhibitors and mutation AN3 CA viability, % of untreated • Erdafitinib-resistant AN3 CA cell lines were generated by gradually beta-Actin concentration, log M 0 mTOR 830 205 1006 610 - erdafitinib + erdafitinib -9 -8 -7 -6 -5 -4 status. concentration, log M increasing the concentration of erdafitinib during cell culture. Figure 3. c-MET is not involved in erdafitinib resistance in AN3 CA resistant cell lines. (A) An FDR ≤ 0.05 FDR > 0.05 • Erdafitinib-resistant lines showed cross-resistance to other FGFR inhibitors. increase in c-MET mRNA expression is observed by qPCR analysis. Expression levels were normalized for A B MYC targets, variant 1 expression of GAPDH, beta-actin, HPRT1 and RPS18, averaged and related to the parental line. Cell lines TNFA signaling via NF_B • Multiple factors were identified which may contribute to the resistance to MYC targets, variant 2 Hs 746T and SNU-5 have high expression of c-MET and serve as reference for expression. (B) Increased cell cycle progression: E2F targets erdafitinib: programmed cell death; caspase pathway proliferation prolifera expression of c-MET is not confirmed in protein expression by immunoblot. (C) In vitro drug sensitivity allograft rejection PREDI CT CHEM OTHERAPY SENSITI VITY OF OVARIAN CANCER PATI ENTS Clonetumor 316 nMtumor 948 assaynMtumor 4800assay nM androgen response analysis shows unaltered response of resistant lines to c-MET inhibitor crizotinib when compared to the cell cycle progression: G2/M checkpoint cells cells cells • Missense mutations in FGFR1 and KRAS were found in increasing 1 TGF beta signaling parental line, indicating c-MET is not the main factor responsible for resistance to erdafitinib. epithelial mesenchymal transition A PRECISION MEDICINEA PRECI APPROACH SI ON M EDI TO CI PREDICT NE APPROACH TO PREDI CT SENSITIVITY CHEM OTHERAPY OF OVARIAN SENSITI CAN VIT frequency after selection on higher concentrations of erdafitinib. proliferation drug unfolded protein response; ER stress Resist_04 biosynthesis of bile acids assay sensit ivit y targeted exome targeted late estrogen response c-Myc was detected to have an altered stability in erdafitinib-resistant ascit es ascit es ascit es sequencing sequenc KRAS signaling, upregulated genes • isolat ion isolat ion isolat ion proliferation prolifera blood coagulation cascade Table 1. Mutations in FGFR1 and KRAS were detected by next generation sequencing. FGFR1 and KRAS interferon alpha response lines resulting in differential expression of c-Myc target genes. Clonetumor 4800 tumornM 7200assay nM assay heme metabolism and erythroid lineage Parental targetedcellsimmune exome cells cellsimmunegene cells immune cells mutations were not detected in the parental AN3 CA cell line, while an increase in variant frequency was observed Notch signaling 2 complement cascade sequencing mutation flow cytometry flow cyto in resist_02 and subsequent resistant line resist_05, indicating selection for the mutation occurs. • BET Bromodomain inhibitors, that indirectly target c-Myc, are of potential AN3 CA glycolysis and gluconeogenesis

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 Resist_02 Resist_05 Normalized Enrichment Score targeted exome targeted Variant frequency interest as therapeutic agents to overcome resistance to erdafitinib. Gene Mutation Figure 1. Schematic representation of the in vitro selection of resistant cell lines. The parental Parental Resist_02 Resist_05 Resist_04 Figure 4. Altered expression of Myc target genes revealed by RNA sequencing. (A) The parental FGFR1 M637V - 18.7 38.0 1.9 cell line was cultured with starting concentrations of erdafitinib based on drug sensitivity analysis. The and resist_05 cell line were subjected to RNA sequencing and subsequent gene set enrichment analysis References: Lau et. al. (2019) Cancer Drug Resistance 2:568-579; Zhou et. al. (2020) Journal of concentration erdafitinib with which the cell lines were cultured is indicated. KRAS A146V - 15.8 39.0 - was performed to detect enriched gene sets. (B) Enrichment plot for Myc target gene set V1. Cancer 11:2000-2007; Yue et. al. (2021) Journal of Hematology & Oncology 14:23

NETHERLANDS TRANSLATIONAL RESEARCH CENTER B.V.