Novel Therapeutic Insights in Dedifferentiated Liposarcoma: A Role for FGFR and MDM2 Dual Targeting

Abstract

We aimed to evaluate the therapeutic potential of the pan-FGFR inhibitor erdafitinib to treat dedifferentiated liposarcoma (DDLPS). FGFR expression and their prognostic value were assessed in a series of 694 samples of well-differentiated/dedifferentiated liposarcoma (WDLPS/DDLPS). The effect of erdafitinib-alone or in combination with other antagonists-on tumorigenicity was evaluated in vitro and in vivo. We detected overexpression of FGFR1 and/or FGFR4 in a subset of WDLPS and DDLPS and demonstrated correlation of this expression with poor prognosis. Erdafitinib treatment reduced cell viability, inducing apoptosis and strong inhibition of the ERK1/2 pathway. Combining erdafitinib with the MDM2 antagonist RG7388 exerted a synergistic effect on viability, apoptosis, and clonogenicity in one WDLPS and two DDLPS cell lines. Efficacy of this combination was confirmed in vivo on a DDLPS xenograft. Importantly, we report the efficacy of erdafitinib in one patient with refractory DDLPS showing disease stabilization for 12 weeks. We provide evidence that the FGFR pathway has therapeutic potential for a subset of DDLPS and that an FGFR1/FGFR4 expression might constitute a powerful biomarker to select patients for FGFR inhibitor clinical trials. In addition, we show that combining erdafitinib with RG7388 is a promising strategy for patients with DDLPS that deserves further investigation in the clinical setting.

Keywords: FGFR; JNJ42756493; MDM2; RG7388; liposarcoma.

Figure 1

FGFR1 and FGFR4 expression in WDLPS/DDLPS clinical tumor samples and prognostic values. (A,B) Immunohistochemical analysis in our series of 418 tumors. (A) Representative illustrations of WDLPS cases positive for FGFR4 (left-hand side); DDLPS cases positive for FGFR4 (middle column) and DDLPS cases positive for both FGFR1 and FGFR4 expression (right-hand side) (localization: membrane and cytoplasm, magnification ×100). (B) Box plot showing the distribution of IHC scores for the overexpressing cases* for FGFR1 and FGFR4 in the WDLPS and DDLPS cases of the IHC cohort. (C–F) Prognostic value of FGFR1 and FGFR4 expression in our cohort of 358 WDLPS/DDLPS patients: Kaplan-Meier survival analyses of FGFR1 and/or FGFR4 expression shows correlation between FGFR1 and/or FGFR4 expression and disease-free survival (DFS, C and E) and overall survival (OS, D and F). (G) Effects of treatment with the erdafitinib pan-FGFR inhibitor (JNJ42756493) in a patient with refractory DDLPS. CT scan images of a 57-year-old male patient with a refractory DDLPS before and after erdafitinib treatment. The patient was resistant to doxorubicin, regorafenib, high-dose ifosfamide, trabectedin and presented a stable disease during 12 weeks (according to the RECIST criteria) under erdafitinib treatment. This tumor co-expressed FGFR1 (score 120) and FGFR4 (score 60).

Figure 2

Analysis of the FGFR/FRS2 signaling pathway after FGFR stimulation and inhibition in WDLPS and DDLPS cell lines. Western blot analysis showing (A) The level of expression for the indicated proteins in one WDLPS (93T449) and two DDLPS (IB111 and IB115) cell lines was compared to the level of expression for the same proteins in a soft tissue leiomyosarcoma cell line (IB136) used as a negative control (non-adipose soft tissue sarcoma). (B) The level of phosphorylated and total protein for the indicated proteins in 93T449, IB111 and IB115 cells treated for 30 min with JNJ42756493 at the IC50 concentrations: 0.36 µM (93T449), 1 µM (IB111), and 2 µM (IB115), with or without FGF1 stimulation (10 ng/mL for 15 min) as indicated. (C) The level of phosphorylated and total protein for the indicated proteins in 93T449, IB111 and IB115 cells treated for 15 min and 24 h with JNJ42756493 at the IC50 concentrations: 0.36 µM (93T449), 1 µM (IB111), and 2 µM (IB115). In all Western blot experiments, β-tubulin was used as a loading control. (D) Quantification of the Western blot analyses shown in (C) performed on cells treated with JNJ42756493 for 15 min, 4 h, 24 h, and 72 h.

Figure 3

Effects of RG7388 as a single agent on activation of the downstream signaling pathways. (A) Western blot analysis showing the level of expression for the indicated proteins in 93T449, IB111, IB115 and IB136 cells. (B) Western blot analysis showing the level of phosphorylated and total protein for the indicated proteins in 93T449 cells treated for 15 min, 4 h, 24 h and 72 h with RG7388 at the IC50 concentration (0.04 µM). (C) Quantification of the western blot analyses shown in (B) and in the two other cell lines (IB111 and IB115) performed on cells treated with RG7388 for 15 min, 4 h, 24 h, and 72 h.

Figure 4

Effects of RG7388 in combination with JNJ42756493 on activation of the downstream signaling pathways. (A) Western blot analysis showing the level of phosphorylated and total protein for the indicated proteins in 93T449, IB111 and IB115 cells treated for 15 min, 24 h and 72 h with JNJ42756493 at 0.36 μM (93T449), 1 μM (IB111), and 2 μM (IB115) and RG7388 at 0.04 μM (93T449), 0.2 μM (IB111), and 0.05 μM (IB115). (B) Quantification of the western blot analyses shown in (A) performed on cells treated with JNJ42756493 + RG7388 for 15 min, 4 h, 24 h, and 72 h. (C) Isobologram representation for the 93T449, IB111 and IB115 cell lines. The combination index (CI) values for the JNJ42756493 + RG7388 combination were calculated and were the following: 0.18, 0.40 and 0.26 respectively, indicating synergy for both WDLPS and DDLPS cells.

Figure 5

Effects of JNJ42756493 and RG7388, as single agents and in combination on in vitro tumorigenicity. (A) Effects on apoptosis induction. Quantification of apoptotic cells using annexin V-FITC and propidium iodide staining followed by flow cytometry after 72 h of treatment by JNJ42756493 or RG7388 and in combination (shown in Figure S4A). JNJ42756493 and RG7388 concentrations were determined according to IC₅₀ values, respectively: 0.36 μM and 0.04 μM (93T449), 0.2 μM and 0.04 μM (for the IB111 cells, because of a high level of apoptosis, concentrations were decreased to 1/5 of IC₅₀ values for each single agent) and 2 μM and 0.05 μM (IB115). (B) Effects on cell cycle distribution. Quantification of cell cycle analysis (shown in Figure S4B) using propidium iodide incorporation and flow cytometry after 48 h of treatment using the same drug concentrations on the same cell lines as in (A). (C,D). Effects on long-term survival. Clonogenic assays in the WDLPS (93T449) and DDLPS cell lines (IB111, IB115) and in the control soft tissue LMS cell line (IB136), using the indicated JNJ42756493 and RG7388 concentrations. (C) Representative pictures of the clonogenic assay. (D) Quantification of the number of colonies obtained in the same experimental conditions as in (C).

Figure 6

In vivo evaluation of the combination of JNJ42756493 and RG7388. (A) Tumor growth in mice injected with DDLPS IB115 cells and treated with either vehicle, RG7388, JNJ42756493, or a combination of the two drugs (combination). (B) Kaplan–Meier survival curves for the four cohorts. A log-rank (Mantel–Cox) test was used to calculate p-values comparing the survival rates.