Failures in preclinical and clinical trials of c-Met inhibitors: evaluation of pathway activity as a promising selection criterion

C-Met is a frequently overexpressed or amplified receptor tyrosine kinase involved in metastatic-related functions, including migration, invasion, cell survival, and angiogenesis. Because of its role in cancer progression and metastasis, many inhibitors have been developed to target this pathway. Unfortunately, most c-Met inhibitor clinical trials have failed to show significant improvement in survival of cancer patients. In these trials tumor type, protein overexpression, or gene amplification are the primary selection criteria for patient inclusion. Our data show that none of these criteria are associated with c-Met pathway activation. Hence, it is conceivable that the majority of c-Met inhibitor clinical trial failures are the consequence of a lack of appropriate patient selection. Further complicating matters, c-Met inhibitors are routinely tested in preclinical studies in the presence of high levels of exogenous Hepatocyte Growth Factor (HGF), its activating ligand. In our studies, several tumor cell lines showed sensitivity to a c-Met inhibitor at high HGF concentrations (50 ng/mL). However, when the tumor lines were tested at HGF levels typically detected in human serum (0.4 to 0.8 ng/mL), inhibitor activity was lost. Thus testing c-Met inhibitors at non-physiological concentrations of HGF may lead to incorrect predictions of drug efficacy in vivo.

All media was supplemented with 10% fetal bovine serum (Gibco) and 1% Pen-Strep (Gibco). All cell lines are maintained in a humidified 5% CO2 chamber at 37° C.

Tumor homogenates
Six to eight week old female C3H/HeJ mice (Jackson) were injected intramuscularly with 10 5 KHT murine fibrosarcoma cells. Beginning 5 days after injection, leg-bearing tumors were measured daily using a ruler with punched with holes of a specified size. Mice were euthanized when tumors reached a size of 150-950 mm 3 . Tumors were rinsed in PBS, weighed, and minced into 1-2 mm pieces. Tumors were placed in a tube with a small volume of PBS and homogenized. After homogenization, an equal volume of Cell Lysis Buffer 2 (R&D) was added, for a final volume of PBS/buffer at 4-8 times the volume of the tissue. After the addition of lysis buffer, the sample was allowed to incubate for 30 minutes at room temperature on a rocker. The sample was centrifuged at 12,000 rpm for 10 minutes at 4° C and the aqueous layer was removed, aliquoted, and stored at -20° C until analysis by ELISA (R&D).

Microdialysis
Five to six week old female C3H/HeJ mice (Jackson) were injected with 2 × 10 4 or 2 × 10 5 KHT fibrosarcoma tumor cells intramuscularly into the left hind leg. Tumors were measured daily till reaching a volume of 1000 mm 3 . Microdialysis probes were primed with T1 buffer (CMA) at 2 μL/minute prior to insertion. Anesthesia was induced using 3% isoflurane and maintained using 1.5% isoflurane. A single probe was inserted into the center of the tumor, and a second probe was inserted into the healthy muscle of the opposite leg with the use of splittable guide cannulas. The probes were allowed to rest in the tissue for 30 minutes, then flushed at a flow rate of 1 μL/minute using an infusion pump and a peristaltic pump (Harvard Apparatus). 120 μL of dialysate was collected at a flow rate of 1 μL/minute, and frozen at -20° C until analysis by ELISA (R&D). The concentration of HGF was calculated using the value determined in the retrodialysis calibration.

Retrodialysis calibration
Microdialysis probes (CMA) were used to sample HGF concentrations within a tumor of a live mouse.
Microdialysis probes are composed of a membrane designed to allow the exchange of solutes to permit sampling of proteins in the extracellular space within a tumor microenvironment. Relative recovery approaches 100% as the flow rate approaches zero. The flow rate of 1 μL/minute was determined to be the fastest flow rate that provided a resulting dialysate protein concentration well within the assay sensitivity. Probes were calibrated for HGF recovery using the following in vivo retrodialysis equation: R = (C perfusate -C dialysate ) = C perfusate .
Five to six week old female C3H/HeJ mice (Jackson) were injected with 2 × 10 5 KHT fibrosarcoma tumor cells intramuscularly into the left hind leg, and allowed to grow to a size of 1000 mm 3 . Various quantities of HGF were diluted in T1 buffer (CMA) for use as the perfusate. Probes were primed at a flow rate of 2 μL/minute prior to insertion. Anesthesia was induced using 3% isoflurane and maintained using 1.5% isoflurane. The probe was inserted into the tumor with a splittable guide cannula. After probe insertion, the tissue was allowed to rest for 30 minutes, then flushed at a flow rate of 1 μL/minute using an infusion pump and a peristaltic pump (Harvard Apparatus). The resulting dialysate, collected at a flow rate of 1 µL/minute, was analyzed via ELISA (R&D) for retrodialysis calculation. The resulting value of 13.12% was used to determine relative recovery of HGF.