Amplification of the MET receptor drives resistance to anti-EGFR therapies in colorectal cancer

Abstract

EGF receptor (EGFR)-targeted monoclonal antibodies are effective in a subset of metastatic colorectal cancers. Inevitably, all patients develop resistance, which occurs through emergence of KRAS mutations in approximately 50% of the cases. We show that amplification of the MET proto-oncogene is associated with acquired resistance in tumors that do not develop KRAS mutations during anti-EGFR therapy. Amplification of the MET locus was present in circulating tumor DNA before relapse was clinically evident. Functional studies show that MET activation confers resistance to anti-EGFR therapy both in vitro and in vivo. Notably, in patient-derived colorectal cancer xenografts, MET amplification correlated with resistance to EGFR blockade, which could be overcome by MET kinase inhibitors. These results highlight the role of MET in mediating primary and secondary resistance to anti-EGFR therapies in colorectal cancer and encourage the use of MET inhibitors in patients displaying resistance as a result of MET amplification.

Figure 1. Whole exome analysis reveals increased MET copy number in CRC samples from patients who developed resistance to anti-EGFR treatment

A–C left side. Whole exome gene copy number analysis of colorectal tumor samples from three patients taken before (in blue) and after (in red) therapy with the EGFR targeted monoclonal antibodies panitumumab (Pmab, patients 1 and 2) or cetuximab (Cmab, patient 3). Individual chromosomes are indicated on the x-axis. The lines indicate the sequencing depth as copy number values relative to a diploid exome (y-axis) over windows of 500,000 base pairs. A–C right side MET amplification was confirmed in the paired tumor samples by quantitative PCR gene copy number analysis. GCN= Gene Copy Number; MAD1L1: reference gene on Chr.7; NGS= next generation sequencing.

Figure 2. MET amplification is selected for in KRAS wild-type CRC samples from patients who developed resistance to anti-EGFR treatment

FISH analysis showing amplification of the MET gene in formalin-fixed paraffin embedded tissue sections from three CRC patients (#1, #2 and #3, A–B–C) who developed resistance to therapy with the anti-EGFR monoclonal antibodies panitumumab or cetuximab. MET (7q31) specific probe is labeled with Texas red, while chromosome 7 centromeric probe D7Z1 (7p11.1-q11.1) is marked in green. Immunohistochemical staining for MET is shown in the left side of each panel. FISH, Original magnification 60x. IHC, original magnification 40x. Pmab= Panitumumab; Cmab= Cetuximab.

Figure 3. Monitoring MET amplification in circulating tumor DNA during anti-EGFR therapy

A. Size of liver metastases at segment 1 (S1, red), segment 2–3 (S2–S3, green) and segment 5 (S5, violet) and of lymph node target lesions at the hepatoduodenal ligament (cyan) during panitumumab therapy in patient #2 at the indicated time points, showing response to panitumumab (Pmab) followed by progression. B. Carcinoembryonic antigen (CEA) levels (blue line) and number of Genome Equivalent (GE, red line) obtained from 1 mL of plasma from patient #2, as assessed by Real Time PCR. C. DNA electrophoresis of PCR products using primers designed to detect the presence of the MET associated amplified rearrangement on Chromosome 7. The lower band corresponds to an 89 bp tumor-specific PCR product which is positive only when the re-arrangement is present. A control assay detecting the wild-type locus generated amplicon of 124 bp (upper band) is also shown.

Figure 4. MET amplification is associated with lack of response to cetuximab in a series of CRC patient derived xenografts (‘exenopatients’)

A. Quantitative PCR gene copy number analysis of MET amplification in a series of cetuximab-resistant ‘xenopatients’, which did not carry genetic alterations in genes previously associated with resistance to anti-EGFR therapy (KRAS, BRAF, NRAS, PIK3CA, HER2). Dotted line indicates an estimated copy number of 2. B. FISH and IHC analysis of MET in CRC samples that showed increased MET copy number by qPCR analysis. Patients (left) and corresponding xenopatients (right) are shown. MET (7q31), red; D7Z1 chromosome 7 centromeric probe (7p11.1-q11.1), green. FISH, Original magnification 60x. IHC, original magnification 40x. C. Growth curves in mice cohorts derived from MET-amplified xenopatients, treated with placebo (blue) or cetuximab 20 mg/Kg i.p. twice a week (red). n = 6 for each treatment arm. Arrows indicate treatment start. D. Prevalence of MET amplification in unselected metastatic colorectal cancer samples, according to q-PCR experiments (left), and in cetuximab-resistant (Cmab), genetically selected (without genetic alterations in KRAS, BRAF, NRAS, PIK3CA, HER2) xenopatients (right). P values were calculated by Fisher’s Exact test.

Figure 5. MET activation confers resistance to cetuximab or panitumumab in colon cancer cell lines in vitro and in vivo

A. DIFI and LIM1215 cell lines were transduced with the following lentiviral vectors: MOCK (empty vector), MET, MET KD (kinase dead) or KRAS, then seeded in 96 wells costars and cultured 7 days in the presence of vehicle (NT), cetuximab (Cmab,1 μg /ml) or panitumumab (Pmab, 1 μg /ml), with or without the MET inhibitor JNJ38877605 (250 nM). B. DIFI and LIM1215 cells were treated for 1 week with increasing concentrations of cetuximab or panitumumab, with or without 20 ng/ml of HGF or HGF plus the MET inhibitor JNJ38877605 (250 nM). C. DIFI and LIM1215 cells were pretreated 24 hours with cetuximab (1 μg /ml), panitumumab (1 μg /ml) or JNJ38877605 (250nM), then stimulated for 15 minutes with HGF (80 ng/ml) in the presence of the indicated inhibitors. Whole-cell extracts were then subjected to western blot analysis and probed with the indicated antibodies. D. Wild-type (WT) DIFI cells or DIFI transduced with HGF were subcutaneously injected in NOD-SCID mice. Upon tumor growth, mice were randomized (7 mice per group) and received intra-tumor injection of saline/ cetuximab or intra-tumor injection of cetuximab and oral administration of JNJ38877605. Arrows indicate treatment start.

Figure 6. Patient-derived xenografts bearing MET amplification respond to MET inhibitors

Tumor growth curves in MET-amplified xenopatients displaying de novo (A, patient M162 shown in Fig. 4) or secondary (B, patient #3 shown in Fig. 1) resistance to cetuximab. Mice were treated with the indicated modalities. Data are presented as means +/− SEM (error bars) of n=5 (A) or n=6 (B) animals for each treatment arm.