Fibroblast growth factor receptor 2 (FGFR2) is a member of transmembrane receptor tyrosine kinase FGFR family that also consists of FGFR1, FGFR3 and FGFR4 [1,2]. Upon binding of the ligand FGF to the extracellular domain, FGFR undergoes dimerization and transautophosphorylation, leading to the activation of multiple signaling pathways that play critical roles in cell survival and proliferation [1]. Oncogenic FGFR2 activation can be caused by gene amplification, activating mutation, or chromosomal rearrangement and has been implicated in various types of cancers including intrahepatic cholangiocarcinoma (iCCA), endometrial cancer and gastric/gastroesophageal junction cancer [[3], [4], [5], [6]]. For example, oncogenic FGFR2 fusions, which typically fuse FGFR2 exons 1–17 that encode the intact extracellular and kinase domains to a 3′ partner gene that permits protein dimerization, produce a constitutively active protein, and represent the most common FGFR2 alterations in iCCA with a prevalence rate of 10–15 % [7,8].

Several pan-FGFR inhibitors including pemigatinib (1), futibatinib (2), infigratinib (3), and erdafitinib (4) have been approved for the treatment of patients with CCA bearing FGFR2 fusions/rearrangements or urothelial carcinoma bearing FGFR2/FGFR3 genetic alterations (Fig. 1) [[9], [10], [11], [12]]. While these inhibitors provided clinical proof of concept of FGFR2 as a cancer therapeutic target, off-isoform toxicity and on-target acquired resistance limited their utility. For example, inhibition of FGFR1 by these pan-FGFR inhibitors increased phosphate reabsorption in the kidney and resulted in hyperphosphatemia, which was recorded in 55–91 % of patients in phase II clinical trials [[13], [14], [15], [16], [17]]. Moreover, inhibition of FGFR4 can cause diarrhea in 15–36 % of patients [[13], [14], [15], [16],18]. Therefore, development of selective FGFR2 inhibitors is highly desirable.

In 2022, Turner and coworkers described the fragment-to-lead optimization of selective FGFR2 inhibitors [19]. While the representative compound 5 potently inhibited FGFR2 kinase activity with an IC50 value of 29 nM, it was only 13.4-fold selective for FGFR2 over FGFR1 (Fig. 1). In 2023, Subbiah and coworkers reported RLY-4008 (6) as the first highly selective, irreversible FGFR2 inhibitor by covalently targeting Cys491 in the P-loop (Fig. 1) [20,21]. Compound 6 displayed an IC50 value of 3 nM for FGFR2 inhibition and exhibited high selectivity against FGFR1 and FGFR4. Compound 6 also demonstrated potent antitumor efficacy in cancer xenograft models bearing oncogenic FGFR2 alterations and currently is in a phase II study in FGFR2-driven solid tumors. Early clinical data showed that compound 6 produced meaningful clinical activity without clinically relevant hyperphosphatemia or diarrhea.

Despite the promising data of compound 6 shown in clinical trials, the detailed structure-activity relationship (SAR) of compound 6 has not been described. Further, it was shown that compound 6 also potently inhibited the mitogen-activated protein kinase (MAPK) kinase 5 (MEK5) and MAPK interacting serine/threonine kinase 2 (MKNK2), which might cause undesirable side effects. So far no selective FGFR2 inhibitors have been approved for clinical use. Herein, we extensively explored the SAR of compound 6 with preferably introducing moieties of more sp3 carbon character, and synthesized 34 new FGFR2 inhibitors. The representative compound LHQ490 (9d) potently inhibited FGFR2 with an IC50 value of 5.2 nM, and exhibited high selectivity in a panel of 416 kinases. LHQ490 also potently suppressed the FGFR2 signaling pathways, selectively suppressed the proliferation of FGFR2-driven BaF3-FGFR2 cells and cancer cells, and induced apoptosis of FGFR2-driven cancer cells.