Recently developed molecularly targeted therapies such as EGFR inhibitors have notably improved the prognosis of patients with cancer. However, patients with KRAS and BRAF mutations do not currently benefit from these therapies. Here, we aimed to examine potential effects of crenolanib as a new molecularly targeted therapy in colorectal cancer. We used multiple colorectal cancer cell lines to investigate the growth-inhibitory effect of crenolanib and its effect in combination with other cytotoxic agents. Primary cultures of patient-derived organoids (PDO), a model that reflects the heterogeneity of clinical colorectal cancer, were used to further validate the effects of crenolanib. Unlike cetuximab, crenolanib remarkably suppressed ERK and AKT/mTOR pathways in HT29 cells with BRAF mutation and in HCT116 cells with KRAS mutation with corresponding growth-suppressing effects. Additive or synergistic effects were observed in treatments with combination of crenolanib and other cytotoxic drugs. Moreover, crenolanib suppressed the expression of stem cell markers, such as OCT4, NANOG, and SOX2. These observations were substantiated in seven PDOs with KRAS mutation and two PDOs without KRAS/BRAF mutations, with crenolanib suppressing the growth of all PDOs regardless of their KRAS mutation status. Furthermore, crenolanib abrogated PDGF- and TGFβ-induced increase of OCT4-positive cells in PDOs. Together, these findings suggest that crenolanib may have clinical utility for patients with colorectal cancer, especially patients with KRAS/BRAF mutations.

Implications:

These findings indicate that crenolanib can be a useful target agent for patients with colorectal cancer, especially patients with KRAS/BRAF mutations.

Cancer incidence continues to increase worldwide, and colorectal cancer is a frequent cause of cancer‐related deaths (1). Drugs for new therapeutic targets in colorectal cancer have been developed, and treatment outcomes have improved (2). Tumor growth, migration, invasion, and drug resistance are known to be enhanced by various growth factors, such as VEGF (3), platelet-derived growth factor (PDGF; ref. 4), EGF (3), insulin-like growth factor (5), TGFβ (6), and basic fibroblast growth factor (7). Molecular therapies and drugs targeting these growth factor receptors have been developed and have become increasingly common in clinical practice (8, 9). In colorectal cancer, the antibodies bevacizumab, targeting VEGF, and cetuximab and panitumumab, targeting EGFR, have been shown to improve the prognosis of patients with metastases when combined with chemotherapy; these antibodies are widely used (2). However, EGFR inhibitors, such as cetuximab and panitumumab, do not improve the prognosis of patients with colorectal cancer harboring RAS and BRAF mutations.

Recently, a combination of EGFR, BRAF, MEK inhibitors have been demonstrated to provide better therapeutic effects than a single-molecular target drug for patients with BRAF mutations (10); however, in current clinical practice, patients with RAS mutations can only receive a combination therapy with bevacizumab (11). RAS mutations have been observed in about 50% of patients with colorectal cancer with unresectable metastases (2), and the development of the new combination therapy with a new molecular target drug is desired.

PDGF modulates cancer growth and migration through its interaction with PDGF receptors (PDGFR; ref. 4). PDGFRs are cell-surface type III tyrosine kinase receptors of two subtypes, PDGFRA and PDGFRB (12). PDGFRs play an important role in the proliferation and metastasis of various tumor cells by activating the PDGF signaling pathway (13–15). We recently reported that PDGFRB expression in colorectal cancer was an independent predictor of cancer recurrence and that knockdown of PDGFRB reduced colorectal cancer cell growth and invasiveness in vitro (16). Thus, PDGFR has been implicated as a new therapeutic target in colorectal cancer.

Crenolanib is a tyrosine kinase inhibitor that targets tyrosine kinase receptors, including PDGFRA, PDGFRB, and FMS-like tyrosine kinase-3 (FLT3). It is an oral medicine and most widely used in acute myeloid leukemia (AML) with FLT3 mutations (17). Mutations in the FLT3 gene were observed in 30% of all AML cases, and FLT3 inhibitors, including crenolanib, have been used to improve the prognosis of patients with AML (18, 19). However, only 1.9% of patients with colorectal cancer carry mutations in the FLT3 gene (20), and FLT3 inhibitors are not clinically used for colorectal cancer. In addition to AML, due to its effects, crenolanib is gaining attention as a drug that suppresses PDGFR signaling, and it was shown to suppress the proliferation of several cancers, such as non–small cell lung cancer (21) and gastrointestinal stromal tumors (22). Crenolanib also suppressed the activation of PDGFR-alpha/AKT-mediated signaling in oral squamous cell carcinoma (23).

Herein, we aimed to investigate the effect of crenolanib in colorectal cancer cells and explore potential molecular mechanisms of its actions in colorectal cancer. We examined the effects of crenolanib using primary colorectal cancer cells as a clinical patient model and assessed the efficacy of crenolanib in combination with other anticancer drugs to verify its clinical usefulness.

Culture of colorectal cancer cell lines

Colorectal tumor cell lines HCT116, DLD1, RKO, and Caco2, gifted by Dr Bert Vogelstein (Johns Hopkins University), HT29 (EC91072201 and ECACC), and SW480 (EC87092801 and ECACC) were cultured in DMEM supplemented with 10% FBS (Thermo Fisher Scientific), 1% GlutaMAX‐I (Thermo Fisher Scientific Inc.) and 1% penicillin/streptomycin/amphotericin B (Wako Pure Chemical Industries). Cells were kept at 37°C in a humidified atmosphere of 95% air and 5% CO2. Cells were harvested using 0.05% Trypsin-EDTA (Thermo Fisher Scientific Inc.) for passaging and further analyses. Except otherwise indicated, cells were treated with 1 ng/mL PDGF (P-8147, Thermo Fisher Scientific), added to the medium in the examinations. Details of the culture medium used for starvation and experiments are summarized in Supplementary Table S1. The number of cell line passages used in the experiments was 7 to 10 passages for HCT116, 13 to 17 for DLD1, 15 to 18 for RKO, 46 to 49 for Caco2, 143 to 147 for Ht29, and 10 to 14 for SW480. Testing for Mycoplasma infection was performed using the MycoAlert Mycoplasma Detection Kit (Lonza) and the final inspection date was November 6, 2020.

Clinical tissue samples

Clinical tissue samples for establishing patient-derived organoids (PDO) were obtained from patients who underwent primary colorectal cancer resection at the Osaka International Cancer Institute (OICI) from 2015 to 2019. None of the patients received chemotherapy or radiotherapy before surgery. The Review Board and Animal Research Committee of the Osaka International Cancer Institute approved the present study, and written informed consent for the study was obtained from all participants.

Culture of primary colorectal cancer cells

Primary cells (603iCC, 25DiCC, 1M29FiCC, 622iCC, 821iCC, 724iCC, 511iCC, 516iCC, 524iCC, 7407iCC, 509iCC, 614iCC, and 311iCC) obtained from patients' tumors were established according to a previous report and cultured in 2 mL of culture medium (modified stem cell culture medium, Supplementary Table S2; ref. 24). The cells were cultured in plates into two-dimensional PDOs. PDOs were kept at 37°C in a humidified atmosphere of 95% air and 5% CO2. The medium was changed every 2 or 3 days. PDOs were harvested using Accutase (Nacalai Tesque) for passaging and further analyses. Except otherwise indicated, cells were treated with 1 ng/mL PDGF (P-8147, Thermo Fisher Scientific Inc.), added to the culture medium in the examination. Details of the culture medium used for starvation and experiments are summarized in Supplementary Table S3.

Xenograft model for PDOs

PDO's recapitulated histological characteristics of clinical colorectal cancer were assessed by xenotransplantation of PDO and their histological examination. Accutase-dissociated cells (5 × 105 cells) were suspended in Matrigel (BD Biosciences), and they were subcutaneously transplanted into the dorsal flanks of a 7-week-old male NOD/SCID (CLEA). Each PDO was injected into different mice. Mice were sacrificed when the tumors reached a diameter of 10 mm. Mice were weighed weekly and showed reduced bodyweights. Xenograft tumors were fixed in formalin, processed through a series of graded concentrations of ethanol, embedded in paraffin, and sectioned. Sections were stained with hematoxylin and eosin (H&E).

In vitro proliferation assays for drug sensitivity

Colorectal cancer cell lines (5 × 103 per well) and PDOs (1 × 104 per well) were seeded into 96-well plates and incubated for 48 hours. Subsequently, cells were exposed to oxaliplatin (22600 AMX0983, Nippon Kayaku Co. Ltd.), camptothecin (017–13424, Wako Pure Chemical Industries), 5-FU (068–01401, Wako Pure Chemical Industries, Ltd.), crenolanib (CP-868596, Selleck Chemicals), and cetuximab (Merck Biopharma) for 96 hours. The percentage of viable cells was determined using a cell counting kit solution (CCK-8; Dojindo Molecular Technologies).

Evaluation of crenolanib using a flank xenograft model of colorectal cancer cells

As an evaluation of crenolanib, HCT116 and DLD1 cells were injected into the axillary regions of mice (2×106cells/site). After 3 days, all tumors were confirmed by palpation, and the mice were randomly allocated to the control group (DMSO) or crenolanib group (15 mg/kg). Crenolanib (CP-868596, Selleck Chemicals) was given by intraperitoneal injection at 15 mg/kg daily 5 days per week. As an evaluation of the additional effect of crenolanib on anticancer drugs, HCT116 cells were injected into the axillary regions of mice (2×106cells/site). After 3 days, all tumors were confirmed by palpation, and the mice were randomly allocated to the oxaliplatin group or crenolanib and oxaliplatin group (15 mg/kg). Oxaliplatin (22600 AMX0983, Nippon Kayaku) was given by intraperitoneal injection at 6 mg/kg daily 2 days per week. Crenolanib (CP-868596, Selleck Chemicals LLC) was given by intraperitoneal injection at 15 mg/kg daily 5 days per week. Tumors and mouse body weight were measured twice per week. Tumor volume was calculated using the following formula: size (mm2) = D × d, where D (mm) is the longest tumor axis and d (mm) is the shortest tumor axis.

IHC

After deparaffinization and blocking, sections were incubated with primary anti-CK20 rabbit monoclonal antibody (1:800, CST 13063, Cell Signaling Technology) and anti-PDGFRA mouse monoclonal antibody (25 μg/mL, MAB322–500, R&D systems) or anti-PDGFRB mouse monoclonal antibody (1:50, 610113, BD Transduction Laboratories) overnight at 4°C. Thereafter, the sections were probed with a secondary antibody (MACH 2 double stain polymer detection kit #2, Biocare Medical) for 30 minutes at ambient temperature. The signal was detected using the Warp Red Chromogen Kit (Biocare Medical) and Vectastain Universal Elite kit (Vector Laboratories) in accordance with the manufacturer's protocol. All sections were counterstained with hematoxylin.

qRT-PCR analysis

Total RNA was isolated using an RNA Purification Kit (Qiagen). Quantitative assessment was performed by RT-PCR using 100 nmol/L universal probe libraries, a 0.1×FASTStart TaqMan Probe Master (Roche Diagnostics) for designed primers, iTaq Universal SYBR Green Supermix (Bio-Rad) for commercially available primers, 100 nmol/L primers, and 10-ng cDNA for the cDNA amplification of target genes. PCR was performed with 20 μL of the master mix in each well of a 96-well plate, and signals were detected with the CFX Connect Real-Time PCT Detection System (Bio-Rad). The thermocycler was programmed for 1 cycle at 95°C for 10 minutes, followed by 40 cycles at 94°C for 10 s, 60°C for 20 s, and 72°C for 1 s. cDNAs from NTERA-2 cells were used as positive controls. Primers and universal probe libraries (Roche Diagnostics) are listed in Supplementary Table S4.

RAS and BRAF mutation analysis using BNA real-time PCR mutation detection kit extended RAS

DNA was extracted from PDOs using a DNA Purification Kit (Qiagen). The presence of KRAS, NRAS, and BRAF mutations was analyzed using a BNA Real-time PCR Mutation Detection Kit Extended RAS (Riken Genesis).

RNA seq analysis

Gene expressions were analyzed for 11 PDOs, three normal colonic mucosal tissues, and six colorectal cancer cell lines. Total RNA was prepared using an RNA Purification Kit (Qiagen). Total RNA (1 μg) was used to prepare RNA-seq libraries, using a TruSeq Stranded mRNA Library Prep (Illumina) according to the manufacturer's instructions. Multiplexed libraries were sequenced on an illumina NextSeq with single-end 75-bp sequencing. RNA-seq data were mapped to the hg38 genome release using the bioinformatic pipeline of the illumina Base Space Sequence Hub and the Subio software platform (Subio). Principal component analysis (PCA) was conducted using the OriginPro software (Lightstone Corp.).

Western blotting analysis

Cells exposed to appropriate drugs for 72 hours were harvested and lysed by RIPA lysis buffer. Lysates were centrifuged at 10,000 × g for 20 minutes and the pellet discarded. The protein concentration of lysates was measured using the Quick Start Bradford Protein Assay (Bio-Rad), and equal amounts of 10-μg proteins were loaded into 10% TGX polyacrylamide gels (Bio-Rad). The proteins were electrophoresed at 200 V for 40 minutes and transferred to polyvinylidene difluoride membranes. The primary antibodies used are listed in Supplementary Table S5. iBind Flex Western Device (Thermo Fisher Scientific) was used for immunoblotting according to the manufacturer's instructions. The iBind Western System was run for 3 hours with a horseradish peroxidase–linked secondary antibody (#7074, Cell Signaling Technology) at a dilution of 1:2,000. Protein expressions were visualized using the Luminescent Image Analyzer LAS-3000 (Fujifilm Corporation). All experiments were independently performed in triplicates and representative figures presented.

Flow cytometric analysis

The expression of surface proteins in the collected cells was determined using flow cytometry (FCM). Cells were dissociated with Accutase (NacalaiTesque) and stained with PDGFRA (130–110–292; R&D systems), PDGFRB (130–101–427; R&D systems), EpCAM (326808; BioLegend), CD45 (368504; BioLegend), and CD90 (324246; BioLegend). The relative fluorescence intensities were measured with an SH800 cell sorter (SONY). A dimensionality reduction step in two dimensions was performed using t-distributed stochastic neighbor embedding (t-SNE) to visualize high-dimensional data of cell surface markers expression in a low-dimensional space. Data were analyzed with the FlowJo 10.2 software (FlowJo).

Gene silencing by siRNA inhibition

Colorectal cancer cell lines (HT29 and HCT116) were used. For siRNA inhibition, Stealth siRNA for PDGFRA (VHS40517; Thermo Fisher Scientific), Silencer Select siRNA for PDGFRB (s10242; Thermo Fisher Scientific), and a negative control siRNA (12935–112; Thermo Fisher Scientific) were transfected at a concentration of 20 nmol/L using lipofectamine RNAiMAX (Thermo Fisher Scientific) according to the manufacturer's protocol. Cells were incubated in glucose-free Opti-MEM (Thermo Fisher Scientific) and used for further analysis.

Establishment of OCT4-EGFP cells

The vector, PL-SIN-OCT4-EGFP, a gift from James Ellis (Addgene plasmid # 21319), was used to establish cells expressing EGFP under the OCT4 promoter. The vector was transfected into PDO (603iCC) using Lentiviral High Titer Packaging Mix with pLVSIN (Takara Bio Inc.) according to the manufacturer's protocol. EGFP-positive cells were enriched via sorting using an SH800 cell sorter (SONY), and high OCT4 expression was confirmed in cells expressing EGFP as previously reported (25).

Observation of OCT4-EGFP cells by adding PDGF or TGFβ

603iCC transfected with OCT4-EGFP (1 × 104 per well) were seeded into 96-well plates and incubated for 48 hours in adherent-2D cultures. Cells were subsequently treated with culture medium containing PDGF (P8147, Sigma-Aldrich), TGFβ (62863, Wako Pure Chemical Industries), and/or crenolanib (CP-868596, Selleck Chemicals) for 72 hours. EGFP fluorescence and cell confluence were detected using the IncuCyte live-cell imaging system (Essen BioScience).

Sphere formation assay

603iCC (1 × 105 per well) were seeded into Costar 24-well Clear Flat Bottom Ultra-Low Attachment Multiple Well Plates (3743). Cells were cultured in a culture medium that included TGFβ (62863, Wako Pure Chemical Industries) and/or crenolanib (CP-868596, Selleck Chemicals) for 72 hours. Sphere size and the number of spheres more than 500 μm were counted using the IncuCyte live-cell imaging system (Essen BioScience).

Statistical analysis

Continuous variables are expressed as the mean ± standard error of the mean. The significance of the relationship between gene expression and cell count was analyzed using the χ2 test and Wilcoxon's signed rank-sum test. Heat maps were used to display cluster analysis data. All data were analyzed using the JMP software (SAS Institute), and results were considered statistically significant if P< 0.05.

Study approval

The OICI Review Board and the OICI Animal Research Committee approved this study, and written informed consent for the study was obtained from all participants according to the ethics guidelines of the OICI. All experimental protocols, including human and animal, were in accordance with the guidelines of the OICI and Declaration of Helsinki.

Gene mutation and expression analysis of crenolanib-targeting molecules

We analyzed gene expression in AML in which crenolanib has been clinically used and colorectal cancer. The mutation statuses of AML and colorectal cancer were examined using the TCGA database (20), and cancer genes with 3% or more mutations were plotted in the chromosomal maps (Fig. 1A). In AML, there were 25 mutated genes, and mutations in FLT3, the target of crenolanib, were most frequent; in colorectal cancer, there were 317 mutated genes, such as KRAS, APC, and TP53. The related pathways of mutated genes were examined using DAVID (refs. 26, 27; Fig. 1B). Nine pathways were significantly enriched in AML, including the Ras signaling pathway, which is regulated by crenolanib. However, 28 pathways were significantly enriched in colorectal cancer, and the top 10 pathways are shown, including Ras and PI3K–Akt signaling pathways, which are regulated by crenolanib. Next, mutations and expressions of crenolanib-targeted tyrosine kinase receptors were examined. FLT3 gene mutations were more common in AML than in colorectal cancer, and FLT3 expression levels were also enhanced in AML (Fig. 1C; Supplementary Fig. S1). Conversely, PDGFRA and PDGFRB gene mutations were more common in colorectal cancer than in AML, and their gene expressions were enhanced in colorectal cancer. Thus, we focused on the PDGFRs as targets of crenolanib in colorectal cancers. Furthermore, the expression of PDGFRs has been reported in cells surrounding tumors, such as cancer-associated fibroblasts (CAF; refs. 28, 29). Immunohistochemical analysis showed that both PDGFRA and PDGFRB were expressed in colorectal cancer tissues, including CAFs (Fig. 1D). Flowcytometric analysis of clinical tumors was performed to clarify the character of cells consisted in the tumors using t-SNE (Fig. 1E). t-SNE analysis for cancer cells showed that the expression of PDGFRA correlated with PDGFRB levels in some cells, but not in others (Fig. 1F). Despite the noted variations, the receptors were also expressed in many colorectal cancer cells.

Figure 1.

Gene mutation and expression analysis of crenolanib targeting molecules in clinical tumors. A, Manhattan plots of mutated genes from the TCGA database. Genes with more than 3% mutation are depicted on their chromosomal location. Names of genes with a high mutation rate are shown. B, Enriched pathways in AML and colorectal cancer. Pathways that were significantly enriched (P<0.05) are shown in AML. For colorectal cancer, only the top10 significantly enriched pathways are shown. C, mRNA expressions and mutations of crenolanib-targeting tyrosine kinase receptors. Gene expression levels in individual cases are depicted, and information on gene mutations are indicated by color. D, Immunohistochemical staining of PDGFRA, PDGFRB, and CK20 in colorectal cancer tissues; scale bars, 100 μm. E, A dimensionality reduction step using t-SNE. The analysis of t-SNE revealed a representative population that consists of colorectal cancer tissues, such as cancer cells, stromal cells, and white blood cells. F, The extracted cancer cells were shown. Cells expressing PDGFRA and PDGFRB are shown with warm or cold colors, corresponding to high or low expressions, respectively (N = 5,000).

Figure 1.

Gene mutation and expression analysis of crenolanib targeting molecules in clinical tumors. A, Manhattan plots of mutated genes from the TCGA database. Genes with more than 3% mutation are depicted on their chromosomal location. Names of genes with a high mutation rate are shown. B, Enriched pathways in AML and colorectal cancer. Pathways that were significantly enriched (P<0.05) are shown in AML. For colorectal cancer, only the top10 significantly enriched pathways are shown. C, mRNA expressions and mutations of crenolanib-targeting tyrosine kinase receptors. Gene expression levels in individual cases are depicted, and information on gene mutations are indicated by color. D, Immunohistochemical staining of PDGFRA, PDGFRB, and CK20 in colorectal cancer tissues; scale bars, 100 μm. E, A dimensionality reduction step using t-SNE. The analysis of t-SNE revealed a representative population that consists of colorectal cancer tissues, such as cancer cells, stromal cells, and white blood cells. F, The extracted cancer cells were shown. Cells expressing PDGFRA and PDGFRB are shown with warm or cold colors, corresponding to high or low expressions, respectively (N = 5,000).

Close modal

Effect of crenolanib on colorectal cancer cell lines

The effect of crenolanib was examined in colorectal cancer cell lines with KRAS mutations (HCT116 and DLD1) and BRAF mutations (HT29 and RKO). PDGFRA and PDGFRB mRNA expression levels varied in the different cell lines and seemed to be unrelated to KRAS or BRAF mutations (Fig. 2A). Crenolanib inhibited the proliferation of all colorectal cancer cell lines (Fig. 2B). In vivo, tumor sizes were significantly smaller with crenolanib treatment than with DMSO control in mice subcutaneously injected with cancer cells (Fig. 2C). Moreover, the efficacy of crenolanib, in combination with other commonly used colorectal cancer anticancer drugs, such as irinotecan hydrochloride, 5-fluorouracil, and oxaliplatin, was examined. Representative results in DLD1 are shown (Fig. 2D), and the IC50 values of anticancer drugs were significantly decreased (Fig. 2E). Combination indices in all colorectal cancer cell lines are provided in Table 1, which revealed additive or synergistic effects. Furthermore, in vivo experiments with HCT116 showed that crenolanib significantly suppressed tumor growth in combination with oxaliplatin (Fig. 2F).

Figure 2.

Effects of crenolanib on colorectal cancer cell lines. A, Relative expression levels of PDGFRA and PDGFRB in four colorectal cancer cell lines; (N = 3). B, The curves of cell viability of four colorectal cancer cell lines by the concentration gradient of crenolanib are shown; (N = 4). C, Representative figures of DLD1 and proliferation curves of HCT116 and DLD1 in vivo treated with crenolanib and without crenolanib (control; N = 5). D, Representative concentration–effect relationships illustrating the effects of combinations of anticancer drugs and crenolanib (0 and 1 μmol/L) in DLD1 cells (N = 4). E, IC50 values of anticancer drugs in DLD1 cells with crenolanib (1 μmol/L) and without crenolanib (0 μmol/L). F, Proliferation curves of HCT116 in vivo treated with oxaliplatin and with/without crenolanib; (N = 5). Data are presented as the mean ± standard error of the mean. *, P < 0.05.

Figure 2.

Effects of crenolanib on colorectal cancer cell lines. A, Relative expression levels of PDGFRA and PDGFRB in four colorectal cancer cell lines; (N = 3). B, The curves of cell viability of four colorectal cancer cell lines by the concentration gradient of crenolanib are shown; (N = 4). C, Representative figures of DLD1 and proliferation curves of HCT116 and DLD1 in vivo treated with crenolanib and without crenolanib (control; N = 5). D, Representative concentration–effect relationships illustrating the effects of combinations of anticancer drugs and crenolanib (0 and 1 μmol/L) in DLD1 cells (N = 4). E, IC50 values of anticancer drugs in DLD1 cells with crenolanib (1 μmol/L) and without crenolanib (0 μmol/L). F, Proliferation curves of HCT116 in vivo treated with oxaliplatin and with/without crenolanib; (N = 5). Data are presented as the mean ± standard error of the mean. *, P < 0.05.

Close modal
Table 1.

Combination index of crenolanib and antitumor drugs.

HCT116HT29DLD1RKO
Irinotecan hydrochloride 1.0 0.3 0.7 0.5 
5-fluorouracil 0.7 0.8 0.5 0.4 
Oxaliplatin 0.9 1.1 0.7 1.1 
HCT116HT29DLD1RKO
Irinotecan hydrochloride 1.0 0.3 0.7 0.5 
5-fluorouracil 0.7 0.8 0.5 0.4 
Oxaliplatin 0.9 1.1 0.7 1.1 

Crenolanib regulated ERK and AKT/mTOR signaling pathway

The EGFR/RAS/ERK pathway is an important therapeutic target in colorectal cancer and a target of EGFR inhibitors, such as cetuximab and panitumumab (30). However, oncogenic mutations of KRAS and BRAF activate MEK/ERK and PI3K/AKT signaling, which are not affected by EGFR inhibitors in clinical patients. We found that cetuximab did not suppress ERK/phosphorylated ERK (pERK) or AKT/pAKT in both HT29 with BRAF mutation and HCT116 with KRAS mutation (Fig. 3A). On the other hand, crenolanib strongly suppressed ERK/pERK and AKT/pAKT in a concentration-dependent manner in both cell lines. Recently, the PI3K/AKT/mTOR signaling pathway was shown to activate the stemness and chemoresistance of cancer cells (31). Similarly, the mTOR signaling pathway was also enriched in our mutational gene analysis of colorectal cancer. Cetuximab did not suppress mTOR/pmTOR in HT29 and only slightly suppressed the pathway in HCT116. However, crenolanib suppressed mTOR/pmTOR in a concentration-dependent manner in both cell lines. Previous studies have reported that the expression levels of stem-cell–related genes affect resistance to cancer treatment (32). We specifically assessed stem-related genes, octamer-binding protein 4 (OCT4), Nanoghomeobox (NANOG), and SRY-box transcription factor (SOX2) and found their expression levels remarkably suppressed by crenolanib treatment in HT29 and HCT116 (Fig. 3B). Inhibition of ERK phosphorylation and expression of stem cell markers was also confirmed in in vivo tumors of HCT116 and DLD1 treated with crenolanib (Fig. 3C and D). Suppression of the ERK and AKT/mTOR signaling pathways was also confirmed in the knockdown of PDGFRA and PDGFRB (Fig. 3E). The inhibitory effect was stronger in the double knockdown of PDGFRA and PDGFRB than in the knockdown of either PDGFRA or PDGFRB. Similarly, suppression of stem cell markers was also confirmed in double knockdown (Fig. 3F). Thus, in colorectal cancer, crenolanib targeted PDGFRs and suppressed growth and malignancy.

Figure 3.

Molecular mechanism of colorectal cancer suppressed by crenolanib. A, Representative western blots of mTOR, pmTOR, AKT, pAKT, ERK, and pERK in HT29 and HCT116 treated by crenolanib (2 and 4 μmol/L) and cetuximab (100 and 200 nmol/L). B,OCT4, NANOG, and SOX2 mRNA expression levels normalized against the expression level of GAPDH were shown in HT29 and HCT116 with crenolanib and without crenolanib (wild type, WT; N = 3). C, Representative western blots of mTOR, pmTOR, AKT, pAKT, ERK, and pERK in xenograft tumor of DLD1 and HCT116 treated with crenolanib od without crenolanib (negative control, NC). D,OCT4, NANOG, and SOX2 mRNA expression levels normalized against the expression level of GAPDH were shown in xenograft tumor of DLD1 and HCT116 treated with crenolanib od without crenolanib (negative control, NC; N = 4). E, Representative western blots of mTOR, pmTOR, AKT, pAKT, ERK, and pERK in HT29 and HCT116-transfected siRNAs (siPDGFRA, siPDGFRB, siPDGFRA, and siPDGFRB). F,OCT4, NANOG, and SOX2 mRNA expression levels normalized against the expression level of GAPDH were shown in HT29 and HCT116 with transfection of siRNAs (siPDGFRA, siPDGFRB, siPDGFRA, and siPDGFRB; N = 3). Data are presented as the mean ± standard error of the mean. *, P < 0.05.

Figure 3.

Molecular mechanism of colorectal cancer suppressed by crenolanib. A, Representative western blots of mTOR, pmTOR, AKT, pAKT, ERK, and pERK in HT29 and HCT116 treated by crenolanib (2 and 4 μmol/L) and cetuximab (100 and 200 nmol/L). B,OCT4, NANOG, and SOX2 mRNA expression levels normalized against the expression level of GAPDH were shown in HT29 and HCT116 with crenolanib and without crenolanib (wild type, WT; N = 3). C, Representative western blots of mTOR, pmTOR, AKT, pAKT, ERK, and pERK in xenograft tumor of DLD1 and HCT116 treated with crenolanib od without crenolanib (negative control, NC). D,OCT4, NANOG, and SOX2 mRNA expression levels normalized against the expression level of GAPDH were shown in xenograft tumor of DLD1 and HCT116 treated with crenolanib od without crenolanib (negative control, NC; N = 4). E, Representative western blots of mTOR, pmTOR, AKT, pAKT, ERK, and pERK in HT29 and HCT116-transfected siRNAs (siPDGFRA, siPDGFRB, siPDGFRA, and siPDGFRB). F,OCT4, NANOG, and SOX2 mRNA expression levels normalized against the expression level of GAPDH were shown in HT29 and HCT116 with transfection of siRNAs (siPDGFRA, siPDGFRB, siPDGFRA, and siPDGFRB; N = 3). Data are presented as the mean ± standard error of the mean. *, P < 0.05.

Close modal

Effect of crenolanib on PDOs

The effect of crenolanib was examined in nine colorectal cancer PDOs. Representative histological examinations of these nine PDOs are shown in Fig. 4A. All the PDOs examined contained various histological types, such as well-differentiated adenocarcinoma, moderately differentiated adenocarcinoma, and mucinous carcinoma. PCA analysis of RNA seq showed that gene expression of PDOs was diverse. In addition, PDOs are located in a different location from normal colorectal mucosa in colorectal cancer cell lines (Fig. 4B); that is, PDOs were composed of malignant cells that retained the diverse gene expression of clinical tumors. KRAS mutation was detected in seven PDOs, and NRAS and BRAF mutations were not detected in all nine PDOs. PDGFRA and PDGFRB mRNA expression levels varied in different PDOs and seemed unrelated to KRAS mutation (Fig. 4C). Crenolanib inhibited the proliferation of all the PDOs (Fig. 4D). The IC50 value of crenolanib and PDGFRA or PDGFRB expression levels were negatively but not significantly related (Fig. 4E). Cetuximab showed a remarkable growth-inhibiting effect in KRAS wild-type PDOs, but crenolanib showed no discrimination in its growth-inhibiting effect with KRAS mutation (Fig. 4F). Taken together, these findings suggest that crenolanib can be an effective targeted drug that does not depend on gene mutation in colorectal cancer.

Figure 4.

Effects of crenolanib on patient-derived organoids (PDO). A, Representative histological examinations of xenograft tumors from PDOs. Pathological diagnosis of the parental tumor is indicated in parentheses; scale bars, 100 μm. B, Principal component analysis of six colorectal cancer cell lines (red), three normal colonic mucosal tissues (blue), and 11 PDOs (gray). C,PDGFRA and PDGFRB mRNA expression in nine PDOs. Data are normalized against the expression level of the GAPDH gene; (N = 3). D, The curves of cell viability of nine colorectal cancer PDOs by the concentration gradient of crenolanib are shown; (N = 4). E, Relationships between crenolanib IC50 values and PDGFRAor PDGFRB mRNA expression levels in PDOs are shown. F, Cell survival rates of nine PDOs treated with cetuximab (100 nmol/L) and crenolanib (4 μmol/L) are shown; (N = 4). Data are presented as the mean ± standard error of the mean.

Figure 4.

Effects of crenolanib on patient-derived organoids (PDO). A, Representative histological examinations of xenograft tumors from PDOs. Pathological diagnosis of the parental tumor is indicated in parentheses; scale bars, 100 μm. B, Principal component analysis of six colorectal cancer cell lines (red), three normal colonic mucosal tissues (blue), and 11 PDOs (gray). C,PDGFRA and PDGFRB mRNA expression in nine PDOs. Data are normalized against the expression level of the GAPDH gene; (N = 3). D, The curves of cell viability of nine colorectal cancer PDOs by the concentration gradient of crenolanib are shown; (N = 4). E, Relationships between crenolanib IC50 values and PDGFRAor PDGFRB mRNA expression levels in PDOs are shown. F, Cell survival rates of nine PDOs treated with cetuximab (100 nmol/L) and crenolanib (4 μmol/L) are shown; (N = 4). Data are presented as the mean ± standard error of the mean.

Close modal

Crenolanib regulated stemness of colorectal cancer

Given that crenolanib remarkably suppressed stem cell markers in colorectal cancer cell lines, we examined its effects on cancer stem cells. OCT4+ cells possess stem cell-like properties, demonstrated chemoresistance, and promoted metastasis in colorectal cancer (25, 33). We prepared primary cells and 603iCCs expressing EGFP under the control of the OCT4 promoter by transfection (Fig. 5A; Supplementary Fig. S2) and further analyzed their responses to crenolanib treatments. PDGF increased EGFP expression, but crenolanib significantly suppressed it (Fig. 5B). PDGF also increased cell proliferation, which was significantly suppressed by crenolanib (Fig. 5C). In addition, TGFβ is known to enhance AKT/mTOR signaling pathways and stem cell properties in colorectal cancer (34). We also observed enhanced EGFP expression with TGFβ treatment, which was attenuated by crenolanib (Fig. 5D). Analysis of sphere formation assay further showed that the number significantly increased upon TGFβ but decreased with crenolanib treatments. The size of spheres did not significantly increase upon TGFβ, but it decreased with crenolanib treatments. These results indicate that crenolanib is a new therapeutic drug for targeting OCT4+ cells with stem cell-like properties. The RAS/MEK/ERK pathway and the PI3K/Akt/mTOR pathway are intricately related, and mutation-activated RAS and RAF spontaneously enhance MEK/ERK regardless of EGFR signaling. The BRAF inhibitor suppressed ERK and its negative feedback to EGFR, resulting in activation of the EGFR signaling pathway (35). However, knockdown of PDGFR and crenolanib suppressed the phosphorylation of the ERK and AKT/mTOR pathways. From these results, it was suggested that colorectal cancer depends partly on PI3K/AKT signaling even in the activated state of RAS and RAF, and crenolanib can be a new therapeutic target for colorectal cancer (Fig. 5E).

Figure 5.

Effects of crenolanib on OCT4+ cells. A, Representative image of transfected 603iCC cells; scale bar, 100 μmol/L. B, The curves of EGFP-integrated intensity were shown in OCT4-EGFP–expressing 603iCCs treated with/without PDGF and with/without crenolanib; (N = 9). C, Representative images and the curves of cell confluence normalized to 0 hours were shown in OCT4-EGFP–expressing 603iCCs treated with/without PDGF and with/without crenolanib; (N = 9) scale bars, 200 μmol/L. D, EGFP integrated intensity of OCT4-EGFP–expressing 603iCCs treated with/without TGFb and with/without crenolanib is shown; (N = 9). In sphere formation assay, the number of spheres and the size of spheres upon crenolanib treatment under TGFβ are shown (N = 9). E, Schema of pathways targeted by crenolanib. There are PI3K/AKT/mTOR and RAS/MEK/ERK pathways downstream of EGFR and PDGFR, respectively. The mutations of RAS and BRAF permanently activate RAS and RAF, resulting in the activation of downstream MEK/ERK. Although EGFR inhibition does not suppress ERK phosphorylation, PDGFR inhibition suppresses ERK phosphorylation. The PI3K/AKT/mTOR pathway was also suppressed, and it seems that the phosphorylation of ERK downstream of PDGFR was largely enhanced by PAK and PI3K. Data are presented as the mean ± standard error of the mean; *, P < 0.05.

Figure 5.

Effects of crenolanib on OCT4+ cells. A, Representative image of transfected 603iCC cells; scale bar, 100 μmol/L. B, The curves of EGFP-integrated intensity were shown in OCT4-EGFP–expressing 603iCCs treated with/without PDGF and with/without crenolanib; (N = 9). C, Representative images and the curves of cell confluence normalized to 0 hours were shown in OCT4-EGFP–expressing 603iCCs treated with/without PDGF and with/without crenolanib; (N = 9) scale bars, 200 μmol/L. D, EGFP integrated intensity of OCT4-EGFP–expressing 603iCCs treated with/without TGFb and with/without crenolanib is shown; (N = 9). In sphere formation assay, the number of spheres and the size of spheres upon crenolanib treatment under TGFβ are shown (N = 9). E, Schema of pathways targeted by crenolanib. There are PI3K/AKT/mTOR and RAS/MEK/ERK pathways downstream of EGFR and PDGFR, respectively. The mutations of RAS and BRAF permanently activate RAS and RAF, resulting in the activation of downstream MEK/ERK. Although EGFR inhibition does not suppress ERK phosphorylation, PDGFR inhibition suppresses ERK phosphorylation. The PI3K/AKT/mTOR pathway was also suppressed, and it seems that the phosphorylation of ERK downstream of PDGFR was largely enhanced by PAK and PI3K. Data are presented as the mean ± standard error of the mean; *, P < 0.05.

Close modal

The development of treatments targeting cancer cells has advanced considerably over time, and many cytotoxic drugs and compounds targeting cancer cell-specific molecules have been developed for treating patients with cancer (2). In the last 30 years, the development of small-molecule kinase inhibitors has advanced remarkably, resulting in longer patient survival (36).

Patients with metastatic colorectal cancer with the BRAF V600E mutation have poor prognoses, and a recent study showed that the combination therapy of encorafenib (BRAF inhibitor), cetuximab (EGFR inhibitor), and binimetinib (MEK inhibitor) resulted in significantly longer overall survival and a higher response rate than standard therapy such as cetuximab and irinotecan or cetuximab and FOLFIRI (folinic acid, fluorouracil, and irinotecan; ref. 10). Because the combination of existing molecular-targeted drugs may enhance the therapeutic effect in this way, we examined the effect of crenolanib on colorectal cancer in search of new treatments.

In this study, we examined the pathways through which crenolanib regulates colorectal cancer cells. The stimulation of tyrosine kinase receptors by cell growth factors activates two major pathways; RAS activation, with consequent RAF, MEK, and ERK activation (MAPK signaling pathway), and PI3K activation, followed by AKT activation (PI3K/AKT signaling pathway). Tumor cells with high MAPK activity ceased to proliferate, enhanced EMT, and expressed stem cell markers (37). AKT and ERK play critical roles in cancer, and their imbalanced suppression could reportedly lead to treatment resistance (38). Our results showed that crenolanib strongly suppressed ERK and AKT regardless of the mutation status in KRAS or BRAF. It is known that when ERK is suppressed by a BRAF inhibitor, negative feedback of ERK to EGFR is suppressed, resulting in the activation of EGFR signaling (35). The results of the interim analysis of the BEACON study for BRAF V600E mutant colorectal cancer showed that the prognosis was slightly better when the BRAF inhibitor and EGFR inhibitor were used in combination with the MEK inhibitor, but the additional effect was less than expected (10). Our results showed that even in colorectal cancer with BRAF/KRAS mutation, the phosphorylation of ERK could be suppressed by the PDGFR inhibitor, crenolanib. It was considered that even in the activated state of RAS and RAF, the phosphorylation of ERK depends on stimulation from the PI3K/AKT/mTOR pathway. Because the effects of PI3K/AKT/mTOR inhibitors have been confirmed on colorectal cancer cell lines, and Phase I and II clinical trials of various PI3K/AKT/mTOR inhibitors are ongoing (39). However, it seems that a combination of molecular-targeted drugs that suppress these two major pathways simultaneously, rather than each pathway alone, may be an effective therapeutic method. Also as the limitation of our study, the combination of crenolanib and other molecular-targeted drugs has not been examined. Further study and development of treatment with molecular-targeted therapeutic agents are desired.

In addition, the IC50 value of colorectal cancer cells was between 3.2 and 12.9 μmol/L in this study, which was higher than that of AML cell lines with FLT3 mutations (40). The IC50 value of AML cell lines without FLT3 mutations, melanoma cell lines (41), and breast cancer cell lines (42) was also between 1 and 20 μmol/L. Moreover, clinical cancer exists with the cancer microenvironment and is more complex. Although the tumor-suppressing effect of crenolanib was also shown in PDO as a human tumor model, future studies should include the elucidation of the effects of crenolanib on colorectal cancer with tumor microenvironment and detemininig the optimal dose. Because crenolanib is also known to have anti-angiogenic effects in the cancer microenvironment (43).

Finally, we focused on the inhibitory effect on cancer stem cells. OCT4 is reportedly expressed in embryonic stem cells, maintaining their pluripotency and self-renewal potential (44); OCT4 can enhance treatment resistance by inducing cancer stem cell-like properties and EMT (32). We previously reported that OCT4 expression is correlated with a poor prognosis in patients with colorectal cancer, and OCT4+ colorectal cancer cells have self-renewal and differentiation abilities (25). PDGF and TGFβ are well-known strong EMT and stemness inducers that promote chemoresistance (4, 45), and our current study demonstrated that OCT4 expression enhanced by PDGF and TGFβ was suppressed by crenolanib. Our results suggest that crenolanib can suppress stemness, typified by OCT4 stimulation by various cytokines and growth factors in vivo, and may, therefore, weaken resistance to anticancer drugs.

In conclusion, we showed that crenolanib could suppress cancer growth and stemness by suppressing ERK and AKT. Crenolanib may be a new therapeutic target for colorectal cancer with or without RAS and BRAF mutations.

No disclosures were reported.

S. Fujino: Resources, data curation, funding acquisition, investigation, visualization, writing–original draft. N. Miyoshi: Conceptualization, supervision, funding acquisition, project administration. A. Ito: Resources, data curation. M. Yasui: Resources. M. Ohue: Resources, supervision. T. Ogino: Resources. H. Takahashi: Visualization. M. Uemura: Visualization. C. Matsuda: Resources. T. Mizushima: Supervision. Y. Doki: Supervision. H. Eguchi: Supervision.

This work was supported by the KAKENHI grant numbers 17K16542 and 19K18145, a 38th Japan Medical Woman's Association Academic research grant, and a JSS Young Researcher Award. The results here are in part based upon data generated by the TCGA Research Network: https://www.cancer.gov/tcga. We would like to thank Editage (www.editage.com) for English language editing.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Bray
F
,
Ferlay
J
,
Soerjomataram
I
,
Siegel
RL
,
Torre
LA
,
Jemal
A
. 
Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
.
CA Cancer J Clin
2018
;
68
:
394
424
.
2.
Yoshino
T
,
Arnold
D
,
Taniguchi
H
,
Pentheroudakis
G
,
Yamazaki
K
,
Xu
RH
, et al
Pan-Asian adapted ESMO consensus guidelines for the management of patients with metastatic colorectal cancer: a JSMO-ESMO initiative endorsed by CSCO, KACO, MOS, SSO and TOS
.
Ann Oncol
2018
;
29
:
44
70
.
3.
Bhattacharya
R
,
Fan
F
,
Wang
R
,
Ye
X
,
Xia
L
,
Boulbes
D
, et al
Intracrine VEGF signalling mediates colorectal cancer cell migration and invasion
.
Br J Cancer
2017
;
117
:
848
55
.
4.
Wang
Y
,
Appiah-Kubi
K
,
Wu
M
,
Yao
X
,
Qian
H
,
Wu
Y
, et al
The platelet-derived growth factors (PDGFs) and their receptors (PDGFRs) are major players in oncogenesis, drug resistance, and attractive oncologic targets in cancer
.
Growth Factors
2016
;
34
:
64
71
.
5.
Vigneri
PG
,
Tirro
E
,
Pennisi
MS
,
Massimino
M
,
Stella
S
,
Romano
C
, et al
The insulin/IGF system in colorectal cancer development and resistance to therapy
.
Front Oncol
2015
;
5
:
230
.
6.
Fournier
PG
,
Juarez
P
,
Jiang
G
,
Clines
GA
,
Niewolna
M
,
Kim
HS
, et al
The TGF-beta signaling regulator PMEPA1 suppresses prostate cancer metastases to bone
.
Cancer Cell
2015
;
27
:
809
21
.
7.
Fessler
E
,
Borovski
T
,
Medema
JP
. 
Endothelial cells induce cancer stem cell features in differentiated glioblastoma cells via bFGF
.
Mol Cancer
2015
;
14
:
157
.
8.
Piccart-Gebhart
M
,
Holmes
E
,
Baselga
J
,
de Azambuja
E
,
Dueck
AC
,
Viale
G
, et al
Adjuvant lapatinib and trastuzumab for early human epidermal growth factor receptor 2-positive breast cancer: results from the randomized phase III adjuvant lapatinib and/or trastuzumab treatment optimization trial
.
J Clin Oncol
2016
;
34
:
1034
42
.
9.
Slamon
DJ
,
Neven
P
,
Chia
S
,
Fasching
PA
,
De Laurentiis
M
,
Im
SA
, et al
Phase III randomized study of ribociclib and fulvestrant in hormone receptor-positive, human epidermal growth factor receptor 2-negative advanced breast cancer: MONALEESA-3
.
J Clin Oncol
2018
;
36
:
2465
72
.
10.
Kopetz
S
,
Grothey
A
,
Yaeger
R
,
Van Cutsem
E
,
Desai
J
,
Yoshino
T
, et al
Encorafenib, Binimetinib, and Cetuximab in BRAF V600E-mutated colorectal cancer
.
N Engl J Med
2019
;
381
:
1632
43
.
11.
Watanabe
T
,
Muro
K
,
Ajioka
Y
,
Hashiguchi
Y
,
Ito
Y
,
Saito
Y
, et al
Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2016 for the treatment of colorectal cancer
.
Int J Clin Oncol
2018
;
23
:
1
34
.
12.
Williams
LT
. 
Signal transduction by the platelet-derived growth factor receptor
.
Science
1989
;
243
:
1564
70
.
13.
Steller
EJ
,
Raats
DA
,
Koster
J
,
Rutten
B
,
Govaert
KM
,
Emmink
BL
, et al
PDGFRB promotes liver metastasis formation of mesenchymal-like colorectal tumor cells
.
Neoplasia
2013
;
15
:
204
17
.
14.
Hayashi
Y
,
Bardsley
MR
,
Toyomasu
Y
,
Milosavljevic
S
,
Gajdos
GB
,
Choi
KM
, et al
Platelet-derived growth factor receptor-alpha regulates proliferation of gastrointestinal stromal tumor cells with mutations in KIT by stabilizing ETV1
.
Gastroenterology
2015
;
149
:
420
32
.
15.
Heske
CM
,
Yeung
C
,
Mendoza
A
,
Baumgart
JT
,
Edessa
LD
,
Wan
X
, et al
The role of PDGFR-beta activation in acquired resistance to IGF-1R blockade in preclinical models of rhabdomyosarcoma
.
Transl Oncol
2016
;
9
:
540
7
.
16.
Fujino
S
,
Miyoshi
N
,
Ohue
M
,
Takahashi
Y
,
Yasui
M
,
Hata
T
, et al
Platelet-derived growth factor receptor beta gene expression relates to recurrence in colorectal cancer
.
Oncol Rep
2018
;
39
:
2178
84
.
17.
Stein
E
,
Xie
J
,
Duchesneau
E
,
Bhattacharyya
S
,
Vudumula
U
,
Ndife
B
, et al
Cost effectiveness of midostaurin in the treatment of newly diagnosed FLT3-mutated acute myeloid leukemia in the United States
.
Pharmacoeconomics
2019
;
37
:
239
53
.
18.
Smith
CC
,
Lasater
EA
,
Lin
KC
,
Wang
Q
,
McCreery
MQ
,
Stewart
WK
, et al
Crenolanib is a selective type I pan-FLT3 inhibitor
.
Proc Natl Acad Sci U S A
2014
;
111
:
5319
24
.
19.
Daver
N
,
Schlenk
RF
,
Russell
NH
,
Levis
MJ
. 
Targeting FLT3 mutations in AML: review of current knowledge and evidence
.
Leukemia
2019
;
33
:
299
312
.
20.
Hoadley
KA
,
Yau
C
,
Hinoue
T
,
Wolf
DM
,
Lazar
AJ
,
Drill
E
, et al
Cell-of-origin patterns dominate the molecular classification of 10,000 tumors from 33 types of cancer
.
Cell
2018
;
173
:
291
304
.
21.
Wang
P
,
Song
L
,
Ge
H
,
Jin
P
,
Jiang
Y
,
Hu
W
, et al
Crenolanib, a PDGFR inhibitor, suppresses lung cancer cell proliferation and inhibits tumor growth in vivo
.
Onco Targets Ther
2014
;
7
:
1761
8
.
22.
Heinrich
MC
,
Griffith
D
,
McKinley
A
,
Patterson
J
,
Presnell
A
,
Ramachandran
A
, et al
Crenolanib inhibits the drug-resistant PDGFRA D842V mutation associated with imatinib-resistant gastrointestinal stromal tumors
.
Clin Cancer Res
2012
;
18
:
4375
84
.
23.
Wang
J
,
Cui
R
,
Clement
CG
,
Nawgiri
R
,
Powell
DW
,
Pinchuk
IV
, et al
Activation PDGFR-alpha/AKT mediated signaling pathways in oral squamous cell carcinoma by mesenchymal stem/stromal cells promotes anti-apoptosis and decreased sensitivity to cisplatin
.
Front Oncol
2020
;
10
:
552
.
24.
Fujino
S
,
Ito
A
,
Ohue
M
,
Yasui
M
,
Mizushima
T
,
Doki
Y
, et al
Phenotypic heterogeneity of 2D organoid reflects clinical tumor characteristics
.
Biochem Biophys Res Commun
2019
;
513
:
332
9
.
25.
Fujino
S
,
Miyoshi
N
. 
Oct4 gene expression in primary colorectal cancer promotes liver metastasis
.
Stem Cells Int
2019
;
2019
:
7896524
.
26.
Huang da
W
,
Sherman
BT
,
Lempicki
RA
. 
Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists
.
Nucleic Acids Res
2009
;
37
:
1
13
.
27.
Huang da
W
,
Sherman
BT
,
Lempicki
RA
. 
Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources
.
Nat Protoc
2009
;
4
:
44
57
.
28.
Gril
B
,
Palmieri
D
,
Qian
Y
,
Anwar
T
,
Liewehr
DJ
,
Steinberg
SM
, et al
Pazopanib inhibits the activation of PDGFRbeta-expressing astrocytes in the brain metastatic microenvironment of breast cancer cells
.
Am J Pathol
2013
;
182
:
2368
79
.
29.
Frodin
M
,
Mezheyeuski
A
,
Corvigno
S
,
Harmenberg
U
,
Sandstrom
P
,
Egevad
L
, et al
Perivascular PDGFR-beta is an independent marker for prognosis in renal cell carcinoma
.
Br J Cancer
2017
;
116
:
195
201
.
30.
Misale
S
,
Di Nicolantonio
F
,
Sartore-Bianchi
A
,
Siena
S
,
Bardelli
A
. 
Resistance to anti-EGFR therapy in colorectal cancer: from heterogeneity to convergent evolution
.
Cancer Discov
2014
;
4
:
1269
80
.
31.
Deng
J
,
Bai
X
,
Feng
X
,
Ni
J
,
Beretov
J
,
Graham
P
, et al
Inhibition of PI3K/Akt/mTOR signaling pathway alleviates ovarian cancer chemoresistance through reversing epithelial–mesenchymal transition and decreasing cancer stem cell marker expression
.
BMC Cancer
2019
;
19
:
618
.
32.
Chiou
SH
,
Wang
ML
,
Chou
YT
,
Chen
CJ
,
Hong
CF
,
Hsieh
WJ
, et al
Coexpression of Oct4 and Nanog enhances malignancy in lung adenocarcinoma by inducing cancer stem cell-like properties and epithelial–mesenchymaltransdifferentiation
.
Cancer Res
2010
;
70
:
10433
44
.
33.
Dai
X
,
Ge
J
,
Wang
X
,
Qian
X
,
Zhang
C
,
Li
X
. 
OCT4 regulates epithelial–mesenchymal transition and its knockdown inhibits colorectal cancer cell migration and invasion
.
Oncol Rep
2013
;
29
:
155
60
.
34.
Nakano
M
,
Kikushige
Y
,
Miyawaki
K
,
Kunisaki
Y
,
Mizuno
S
,
Takenaka
K
, et al
Dedifferentiation process driven by TGF-beta signaling enhances stem cell properties in human colorectal cancer
.
Oncogene
2019
;
38
:
780
93
.
35.
Corcoran
RB
,
Ebi
H
,
Turke
AB
,
Coffee
EM
,
Nishino
M
,
Cogdill
AP
, et al
EGFR-mediated re-activation of MAPK signaling contributes to insensitivity of BRAF mutant colorectal cancers to RAF inhibition with vemurafenib
.
Cancer Discov
2012
;
2
:
227
35
.
36.
Bhullar
KS
,
Lagaron
NO
,
McGowan
EM
,
Parmar
I
,
Jha
A
,
Hubbard
BP
, et al
Kinase-targeted cancer therapies: progress, challenges and future directions
.
Mol Cancer
2018
;
17
:
48
.
37.
Blaj
C
,
Schmidt
EM
,
Lamprecht
S
,
Hermeking
H
,
Jung
A
,
Kirchner
T
, et al
Oncogenic effects of high MAPK activity in colorectal cancer mark progenitor cells and persist irrespective of RAS mutations
.
Cancer Res
2017
;
77
:
1763
74
.
38.
Cao
Z
,
Liao
Q
,
Su
M
,
Huang
K
,
Jin
J
,
Cao
D
. 
AKT and ERK dual inhibitors: the way forward?
Cancer Lett
2019
;
459
:
30
40
.
39.
Bahrami
A
,
Khazaei
M
,
Hasanzadeh
M
,
Shahid Sales
S
,
Joudi Mashhad
M
,
Farazestanian
M
, et al
Therapeutic potential of targeting PI3K/AKT pathway in treatment of colorectal cancer: rational and progress
.
J Cell Biochem
2018
;
119
:
2460
9
.
40.
Zimmerman
EI
,
Turner
DC
,
Buaboonnam
J
,
Hu
S
,
Orwick
S
,
Roberts
MS
, et al
Crenolanib is active against models of drug-resistant FLT3–ITD-positive acute myeloid leukemia
.
Blood
2013
;
122
:
3607
15
.
41.
Sabbatino
F
,
Wang
Y
,
Wang
X
,
Flaherty
KT
,
Yu
L
,
Pepin
D
, et al
PDGFR alpha upregulation mediated by sonic hedgehog pathway activation leads to BRAF inhibitor resistance in melanoma cells with BRAF mutation
.
Oncotarget
2014
;
5
:
1926
41
.
42.
Joglekar-Javadekar
M
,
Van Laere
S
,
Bourne
M
,
Moalwi
M
,
Finetti
P
,
Vermeulen
PB
, et al
Characterization and targeting of platelet-derived growth factor receptor alpha (PDGFRA) in inflammatory breast cancer (IBC)
.
Neoplasia
2017
;
19
:
564
73
.
43.
Berndsen
RH
,
Castrogiovanni
C
,
Weiss
A
,
Rausch
M
,
Dallinga
MG
,
Miljkovic-Licina
M
, et al
Antiangiogenic effects of crenolanib are mediated by mitotic modulation independently of PDGFR expression
.
Br J Cancer
2019
;
121
:
139
49
.
44.
Loh
YH
,
Wu
Q
,
Chew
JL
,
Vega
VB
,
Zhang
W
,
Chen
X
, et al
The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells
.
Nat Genet
2006
;
38
:
431
40
.
45.
Gotzmann
J
,
Fischer
AN
,
Zojer
M
,
Mikula
M
,
Proell
V
,
Huber
H
, et al
A crucial function of PDGF in TGF-beta–mediated cancer progression of hepatocytes
.
Oncogene
2006
;
25
:
3170
85
.