[¹⁸ F]FDG-PET imaging is an early non-invasive pharmacodynamic biomarker for a first-in-class dual MEK/Raf inhibitor, RO5126766 (CH5126766), in preclinical xenograft models

The human colon cancer cell lines HCT116, COLO205 and COLO320DM were purchased from the American Type Culture Collection (ATCC). All cells were maintained in the designated media and indicated concentrations of heat-inactivated fetal bovine serum (Gibco) and L-glutamine (Sigma) according to the ATCC recommendations. Cells were grown at 37°C in an atmosphere of 5% CO₂. RO5126766 (CH5126766) was synthesized in Chugai Pharmaceuticals Co., Ltd. For in vitro and in vivo studies, the drug was dissolved in DMSO (Wako Chemicals GmbH) to yield a 2.5 mg/mL stock solution concentration and stored at -20°C. The solutions of RO5126766 used for in vitro and in vivo experiments were freshly prepared on each experimental day. The vehicle and RO5126766 stock solutions were diluted 1:20 with the diluent (10.5% aqueous solution of 2-hydroxypropyl-β-cyclodextrin (Celdex HP- β-CD, HPCD, Sigma)) on each dosing day.

[¹⁸ F]FDG uptake was determined in untreated HCT116, COLO205 and COLO320DM cells as well as treated with vehicle only as a control or RO5126766 at indicated concentrations. 1×10⁵ cells/well were seeded in 6-well plates (Costar®) together with appropriate doses of RO5126766 for indicated times. Cell culture medium was changed to glucose-free and 0.37 MBq of [¹⁸ F]FDG was added to each well and incubated for 1 hour in 5% CO₂ atmosphere at 37°C. The cells were washed three times with ice cold PBS and radioactivity was measured using a 1480 Automatic gamma counter Wizard3 (Perkin Elmer). [¹⁸ F]FDG incorporation was determined and expressed relative to total protein concentration. Protein content was determined using the Thermo Scientific Pierce BCA Protein Assay Kit (Waltham, USA).

Plasma membrane fractionation was performed using a membrane protein extraction kit from BioVision. According to the manufacturer's recommendations, 5×10⁸ HCT116 cells were used for protein extraction. The purity of the plasma membrane protein fraction was assessed by Western blot analysis of the plasma membrane marker (mouse monoclonal antibodies to Na⁺, K⁺-ATPase from Abcam). For Western blot analysis, 15 μg of plasma membrane-associated or of the cytosol proteins were separated on 4-12% poly-acrylamide gels (Invitrogen) and Western blotting was performed. The rabbit polyclonal antibodies to glucose transporters GLUT1 and GLUT3 were from Abram (Cambridge, UK). (Secondary HRP-linked anti-mouse and anti-rabbit IgG were from Cell Signaling (In Vitro Sweden AB, Stockholm, Sweden). Bands were visualized with Western blotting Luminol Reagent (Santa Cruz). The images were captured using a LAS-1000 from Fujifilm or exposed to X-ray film (Fujifilm, Tokyo).

Female athymic nude (nu/nu) mice, age 5–6 weeks (18-22 g) were purchased from Scanbur AB (Sollentuna, Sweden). Animal care, handling and health monitoring were carried out in accordance with the Guidelines for Accommodation and Care of Laboratory Animals. All animal experiments were performed in accordance with protocols approved by the Institutional Animal Care committee.

For the tumor xenografts, 5×10⁶/mouse human cancer colon carcinoma cells were inoculated subcutaneously in the right flank of Balb-nu/nu mice. Once tumors were established (150–200 mm³), mice were randomized into groups with similar mean tumor volumes at the start of the study. Tumor volume and body weight were measured two times per week. Tumor volumes were determined with digital caliper using the formula (tumor length × width ² )/2. Tumor growth inhibition (TGI) was calculated using the following formula: TGI = [1 − (T − T ₀ )/(C − C ₀ )] × 100, where T and T ₀ are the mean tumor volumes on a specific experimental day and on the first day of treatment, respectively, for the experimental groups and likewise, where C and C ₀ are the mean tumor volumes for the control group. The daily administration of RO5126766 was performed orally at doses 0.1, 0.3 and 1.0 mg/kg. The doses were selected based on the results of preliminary studies. The maximal tolerated dose (MTD) was defined as the maximum dose associated with <20% weight loss and no toxic deaths. The MTD in the three xenograft models was 1.5 mg/kg for RO5126766.

MicroPET imaging was performed by standard protocols as described previously [26]. Mice were fasted for 6–8 hours prior to start of imaging session [27]. [¹⁸ F]FDG (obtained as an aliquot from daily clinical productions at Karolinska University Hospital, 7–8 MBq per mouse, maximum volume of 200 μl) was administered to awake, warmed (37°C) mice by a bolus injection via the tail vein. Forty to sixty minutes after the tracer injection, the mice were anaesthetized with isoflurane, controlled by an E-Z anaesthesia vaporizer (5% initially and then 1.5% to maintain anaesthesia, blended with 7:3 air/O2 and delivered through a Microflex non-rebreather mask from Euthanex Corporation, Palmer, PA). The mice were placed on a heating pad (37°C) on the camera bed, with most of the body in the field-of-view (7.68 cm). Emission data were collected for 20 minutes in list mode with MicroPET Focus 120 scanner (CTI Concorde Microsystems). Data were processed using MicroPET Manager (CTI Concorde Microsystems). PET data were acquired in fully three-dimensional (3-D) mode and images were reconstructed by standard 2-D filtered back projection using a ramp filter. The matrix size of the reconstructed images was 128 × 128 × 95 with a spatial resolution of 1.3 mm. Data were corrected for randoms, dead time and decay. Standard Uptake Values (SUV) were calculated for 3D regions of interest (ROI), using Inveon Research Workplace software (Siemens Medical Solutions). Tumor ROIs drawn on the images employed a 75% threshold of the maximum intensity voxel. The ROI counts were normalized to the injected dose and body weight and converted to SUV. The drug effect on tumor metabolism was estimated as%SUV_max change day 1, 2, or 3 compared to day 0 (baseline).

Immediately after the last MicroPET scan, the animals were sacrificed and their tumors were rapidly frozen. Tumor slices (20 μm thickness) were obtained using a cryomicrotome (CM 3050S, Leica Microsystems, Wetzlar, Germany) and were placed on Superfrost Plus microscope slides (Menzel-Glaser, Germany). The sections were placed in a BAS exposure cassette with a 2325 imaging plate (Fujifilm Corporation, Japan) and exposed for at least 1 hour. The quantitative autoradiography data (photo-stimulated luminescence, PSL, unit/mm²) were normalized to the injected dose and body weight.

Statistical significance was examined by Student t test. P-values less than 0.05 were considered a statistically significant difference.

We first investigated in vitro the feasibility of using two RO5126766 sensitive cell lines, HCT116 ( ^G13D K-ras) and COLO205 ( ^V600E B-raf), and one resistant cell line COLO320DM (no K-ras and B-raf mutation) for [¹⁸ F]FDG-PET imaging. The study revealed variations in glucose utilization between the three cell lines. The highest cellular glucose uptake was observed in COLO320DM cells and lowest in COLO205 (Additional file 1: figure S1). The RO5126766 at the concentration of 0.3 μM was a minimal dose demonstrated reduction of ERK/MEK phosphorylation to undetectable levels in HCT116 (K-ras) cells after 2 hours of the treatment start [24]. Variations in [¹⁸ F]FDG uptake in tumor cells exposed to RO5126766 were subsequently examined with 0, 0.3 or 1.3 μM of RO5126766 up to 48 hours. There was no significant change in the cellular accumulation of [¹⁸ F]FDG in the drug-resistant COLO320DM during treatment (Figure 1a). In HCT116 cells, a significant reduction in [¹⁸ F]FDG uptake was observed after 24 h of treatment (67.1%, p < 0.05, Figure 1b) and not at earlier time points. In contrast, in drug-sensitive COLO205 cells, a dose-dependent decrease in [¹⁸ F]FDG uptake was observed as early as after 2 hours of treatment compared to control (54.8%, p < 0.01, Figure 1c).

In order to identify the cellular components determining the uptake and retention of [¹⁸ F]FDG in these cell lines, the expression levels of glucose transporters (GLUTs) and hexokinases were analyzed by Western blotting. GLUT1 was detected in all cell lines. GLUT3 was expressed in COLO320DM and HCT116 but not in COLO205 cells. Hexokinase II was expressed in all cell lines (Figure 1d). To further examine possible mechanisms behind the RO5126766-induced changes in [¹⁸ F]FDG uptake, we used the HCT116 cell line because of its higher basal glucose utilization. We detected significant decreases in the expression of the cellular transmembrane protein GLUT1 in the plasma membrane fraction after 24 hours of treatment with 1.3 μM of RO5126766, compared to the vehicle treated cells. In parallel, an increase of GLUT1 in the cytosol fraction was observed during treatment (Figure 1e). We did not detect significant changes in hexokinase II activity during the treatment of HCT116 cells with 1.3 μM of RO5126766 for 24 hours (Figure 1f).

In vitro experiments demonstrated that RO5126766 treatment resulted in dose-dependent decreases in [¹⁸ F]FDG uptake for both K-ras and B-raf mutants, but not for COLO320DM, the resistant cell line. Furthermore, PET imaging of antitumor activities of RO5126766 and quantification of early response in the three colorectal cancer xenograft models were evaluated on days 0 and 3 of the treatment. Once the tumors were established (approximately 0.2 cm³) and mice were divided into treatment groups (n = 10/group) and treatment was initiated with vehicle and RO5126766 at 0.1, 0.3 or 1.0 mg/kg daily oral gavage for 9 days. RO5126766 treatment (1.0 mg/kg) did not inhibit growth of the COLO320DM tumors (Figure 2a) and these mice were therefore sacrificed after 6 days of treatment when the tumors had reached the size limits allowed by research ethics. In contrast, RO5126766 treatment showed dose-dependent tumor growth inhibition (TGI) in the mice with xenografts of both the mutant models. In HCT116 (K-ras) tumor xenografts the treatment resulted in 80% TGI (0.1 mg/kg), 119% TGI (0.3 mg/kg, p < 0.01) and 157% TGI (1.0 mg/kg, p < 0.01) (Figure 2b). In the COLO205 (B-raf) mutant tumor xenografts TGI’s of 120% (0.3 mg/kg, p < 0.01) and 190% (1.0 mg/kg, p < 0.01) (Figure 2c) were achieved.

[¹⁸ F]FDG uptake was measured in tumors of mice treated with RO5126766 at 0.1, 0.3 or 1.0 mg/kg versus vehicle from day 0 (baseline) to day 3. PET imaging revealed no significant effect on [¹⁸ F]FDG uptake in COLO320DM tumors during the treatment (Figure 2d). In contrast, RO5126766 treatment HCT116 (K-ras) tumors demonstrated significant decrease in metabolic activity on day 3, compared to day 0 (p < 0.05 at 0.3- and 1.0 mg/kg doses and vehicle (p < 0.01). No significant effect was observed in the 0.1 mg/kg dosing group (Figure 2e). Despite the low basal [¹⁸ F]FDG-uptake in COLO205 tumor xenografts, we could detect inhibition by RO5126766 at both 0.3 mg/kg (p < 0.05) and 1.0 mg/kg (p < 0.05) doses. Representative coronal PET images in Figure 2g-i demonstrate the FDG uptake observed in COLO320DM (wt) tumors, HCT116 (K-ras) and COLO205 (B-raf), respectively, at days 0 and 3 of treatment with vehicle versus RO5126766 (only 1.0 mg/kg dose shown).

The HCT116 tumors demonstrated higher basal [¹⁸ F]FDG-uptake and were usually more readily distinguished from background tissues than COLO205 tumors. Therefore, additional PET scans and ex-vivo phosphor imaging for earlier time points than day 3 and one lower dose were performed using HCT116 xenografts. Serial microPET imaging revealed significant decreases in tumor [¹⁸ F]FDG uptake as early as on day 1 after administration of 0.3 mg/kg RO5126766 (Figure 3a, representative coronal images). Maximum reductions of [¹⁸ F]FDG-uptake tumors treated with RO5126766, 0.3 mg/kg on day1, day2 and day3 were 17% (p < 0.01), 19% (p = 0.15) and 35% (p < 0.05) compared with baselines values, respectively (Figure 3b).

In the mice bearing HCT116 xenografts that received the lowest dose RO5126766 (0.1 mg/kg), statistically significant changes in [¹⁸ F]FDG uptake could not be detected by PET on day 3 (Figure 2e). As a complement to PET, ex-vivo phosphor imaging at higher resolution and sensitivity was also performed (Figure 4a,b, representative images). Tumors were sectioned after PET imaging and exposed to phosphor imaging plates together with a blood sample from the same mouse. We observed higher radioactivity concentrations ([¹⁸ F]FDG uptake) in tumors on day 0 compared to day 3 of treatment (Figure 4a). Hematoxylin & eosin examination revealed no difference in examined tumors morphology (Figure 4b). The result was confirmed by statistical analysis of the [¹⁸ F]FDG uptake in a larger population of mice (n = 6, p < 0.01) (Figure 4c). However, the ex vivo technique of course only gave one time point per animal and could not be used for monitoring individual responses over time.

The tumors that were subjected to ex-vivo phosphor imaging and excised from mice treated with 0.1 mg/kg RO5126766 on day 3 were also analyzed histologically (hematoxylin & eosin, H&E) and immunohistochemically (Ki-67) (Figure 5a, b). The difference in [¹⁸ F]FDG uptake in tumors correlated with their proliferative activity as detected with Ki-67 antigen. Figure 5c shows the number of proliferating cells on day 0 (baseline) and day 3 after therapy.