SEARCH

SEARCH BY CITATION

Keywords:

  • NSCLC;
  • EGFR mutation;
  • c-met amplification;
  • T790M;
  • L858R;
  • animal models;
  • crizotinib;
  • erlotinib;
  • TKI-resistance;
  • combination target therapy;
  • cetuximab

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Overall benefits of EGFR-TKIs are limited because these treatments are largely only for adenocarcinoma (ADC) with EGFR activating mutation. The treatments also usually lead to development of resistances. We have established a panel of patient-derived xenografts (PDXs) from treatment naïve Asian NSCLC patients, including those containing “classic” EGFR activating mutations. Some of these EGFR-mutated PDXs do not respond to erlotinib: LU1868 containing L858R/T790M mutations, and LU0858 having L858R mutation as well as c-MET gene amplification, both squamous cell carcinoma (SCC). Treatment of LU0858 with crizotinib, a small molecule inhibitor for ALK and c-MET, inhibited tumor growth and c-MET activity. Combination of erlotinib and crizotinib caused complete response, indicating the activation of both EGFR and c-MET promote its growth/survival. LU2503 and LU1901, both with wild-type EGFR and c-MET gene amplification, showed complete response to crizotinib alone, suggesting that c-MET gene amplification, not EGFR signaling, is the main oncogenic driver. Interestingly, LU1868 with the EGFR L858R/T790M, but without c-met amplification, had a complete response to cetuximab. Our data offer novel practical approaches to overcome the two most common resistances to EGFR-TKIs seen in the clinic using marketed target therapies.

Non-small cell lung cancer (NSCLC) has the highest mortality among all malignancies, whereas it has few effective treatment options. NSCLC are diverse types of diseases including adenocarcinoma (ADC), squamous cell carcinoma (SCC), large cell carcinoma (LCC), adenosquamous carcinoma (PLC), etc. The approval of the first generation reversible inhibitors of epidermal growth factor receptor (EGFR) tyrosine kinase (TKIs), e.g. erlotinib and gefitinib, has delivered significant clinical benefits for a subset of NSCLC patients,1, 2 as confirmed in a recent randomized phase III study (IPASS).3 This subset of patients usually are those of ADC (∼30%),4, 5 instead of SCC (<3%), per studies largely based on Caucasian patient populations. They usually also have EGFR activating mutations, including L858R point mutation and deletion mutations in the tyrosine kinase domain. The patients of East Asian, nonsmoker and women are among the more likely beneficiaries for the higher frequency of these mutations.1, 2 However, majority treatments lead to development of drug resistance.6, 7 The two major mutually exclusive resistant mechanisms are8, 9: T790M “gatekeeper” mutation preventing drug access to the kinase activation site (50%),10 and c-met gene amplification (20%)11, 12 bypassing TKI inhibition through activating alternative signaling pathways such as the ERBB3.13 Therefore, there is urgent and important need of the NSCLC treatment that can overcome TKI resistances. Apparently, existing therapies that were approved for other applications and can also overcome these resistances could be particularly attractive in meeting this need.

Cetuximab is a mouse-human chimeric monoclonal antibody (IgG1) binding to the extracellular domain of EGFR competitively with the EGF ligand, thus an EGFR inhibitor of different mechanism of action from TKIs. Cetuximab are being investigated for treatment of NSCLC in the clinic (FLEX14 and BMS09915) with reported activities in some patients, but has yet to be approved for lack of biomarker to identify the likely responders.

c-MET, or hepatocyte growth factor (HGF) receptor, encodes a heterodimeric transmembrane RTK with HGF as its ligand. Activation of c-MET signaling has been associated with various oncogenic processes, such as invasion, metastasis and angiogenesis. Several genetic alterations of c-MET functionally activate c-MET, including the activating mutations (e.g., papillary renal carcinomas) or gene amplification (e.g., gastric carcinoma and hepatocellular carcinoma). The gene amplification has also frequently been observed in NSCLC after acquiring TKI-induced resistance.11 However, the amplification has also been recently observed in nonexposed NSCLC patients.12 It is believed that the TKI resistance is caused by the activation of alternative pathway such as ERBB3 signaling.16 In particular, mutated or amplified EGFR has also been suggested to drive c-MET activity.17 At present, a number of c-MET inhibitors are being developed for treatment of various cancers. Crizotinib is a newly approved treatment for NSCLC with ALK activation (EML4-ALK fusion) and also a good c-MET inhibitor with actually higher potency against c-MET than ALK.18

Patient-derived xenografts (PDX) (including NSCLC PDX) never manipulated in vitro, mirror patients' histopathology and genetic profiles.19–24 It has improved predictive power as preclinical cancer models, and enables discovery of predictive biomarkers for targeted therapeutics. We have established a large collection of NSCLC PDX models (NSCLC-HuPrime®) of diverse histology subtypes (ADC, SCC, PLC, LCC, etc.) by engrafting naïve patient tumor tissues into immunocompromised mice. Among them, we have identified several models with EGFR activating mutations. Interestingly, some the models are inherently resistant to erlotinib, even though the original patients have never been exposed to erlotinib. The present studies explored tailored treatment strategies to overcome these erlotinib resistances. Our data suggest the marketed crizotinib and cetuximab could potentially be used to treat patients with similar resistances.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Patient tumor samples and engraftment in immunocompromised mice

Freshly and surgically removed tumor tissues were obtained from the patients diagnosed as NSCLC in Hebei Medical University Fourth Hospital through collaboration with Beijing Keluoen Translational Medicine Institute with approval by the Institutional Review Boards of the hospital and the informed consents from patients. The engraftments of patient tumor fragments into immunocompromised mice subcutaneously have been broadly reported.19–21, 25 Briefly, the tumors were sliced into 3 × 3 × 3 mm3 fragments and inoculated subcutaneously on the flank of mice (Balb/c nude, 6- to 8-weeks old female mice, Beijing HFK Bioscience Co., Beijing, China). The tumor growth was monitored twice weekly using a caliper. The established tumor models, called passage 0 or P0, were serially re-engrafted to maintain tumors in vivo. These subsequent passages were called P1, 2, 3… (<10). When tumors sizes reach 500–700 mm3 (1/2 length × width2), they were harvested for the next round of engraftment for serial passage or conducting studies of pharmacology, histopathology, immunohistology, cellular and molecular analysis. All procedures were under sterile conditions at Crown Bioscience SPF facility and conducted in strict accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Committee on the Ethics of Animal Experiments of Crown Bioscience (Crown Bioscience IACUC Committee).

Evaluation of antitumor activity

When tumor volume reaches 100–150 mm3, the mice were randomly grouped into two groups of five mice with similar average tumor volume. Immediately after grouping, the control group was treated with vehicle (PBS, weekly IP injection for 2 weeks), and the treatment groups were treated with one of the following: cetuximab (weekly IP injection for 2 weeks, 1 mg/mouse or 50 mg/kg, Merck KGaA), erlotinib (daily oral, 50 mg/kg, Nanjing Ange Pharmaceutical Co.) and crizotinib (daily oral, 50 mg/kg, Selleckchem.com). The tumor growth was monitored twice weekly, and %ΔTC value were calculated for assessing tumor response to the treatment (ΔT = tumor volume change in the treatment group and ΔC = tumor volume change in control group).

EGFR mutation analysis

EGFR gene hotspot analyses were carried out to identify the mutations in the tumors. Briefly, genomic DNA was extracted from the tissues using DNeasy mini kit (QIAGEN, Valencia, CA) according to the manufacturer's instructions. All DNA samples are PCR amplified in a 50-μL reaction using primers: EGFR-EXON19-F: 5′-GTGCATCGCTGGTAACATCCA-3′; EGFR-EXON19-R: 5′-GGAGATGAGCAGGGTCTAGAGCA-3′; EGFR-EXON20-F: 5′-CGCATTCATGCGTCTTCACC-3′; EGFR-EXON20-R: 5′-CTATCCCAGGAGCGCAGACC-3′; EGFR-EXON21-F: 5′-TGGCATGAACATGACCCTGAA-3′: EGFR-EXON21-R: 5′-CAGCCTGGTCCCTGGTGTC-3′. Polymerase chain reaction was performed in 50-μL reaction mixtures containing: 100 ng of genomic DNA, 5 μL 10× PCR buffer, 0.2 μM each of primers, 0.2 mM 4× dNTPs and 1 μL TaqE. Reaction was carried for 40 amplification cycles. The amplified PCR products were gel purified and sequenced by Sanger Automated Sequencer (ABI). EGFR-specific primers were designed around regions that are found to be frequently mutated in cancer cells, including exons 18, 19, 20 and 21. PCR reactions were set using these primers to amplify individual exons. The specificity of the primers to human genes had been assured by BLAST search, as well as using mouse genomic DNA as a negative control in PCR amplification. Sequencing data alignment analysis and mutation identification was performed using BioEdit software.

IHC analysis of HuPrime® tumors

Standard immunohistochemistry (IHC) was used to analyze tumor tissues from the HuPrime® models. Briefly, the tissues were fixed in 10% neutral buffered formalin and embedded in paraffin per standard histological procedures. After deparaffinization and rehydratation, 3-μm thick tissue sections were pretreated at 95°C in 0.01 M sodium citrate, pH 6.0 solution for 30 min, followed by staining with rabbit anti-human monoclonal pERK or pEGFR antibody (Cell Signaling, Boston, USA). Positive staining was detected using Ultra Vision LP large Volume Detection System HRP Polymer (ready-to-use) Kit (Lab Vision, Fremont, CA). DAB was used as the chromogenic substrate, and sections were counterstained with Gill's hematoxylin (Fisher Scientific, Fair Lawn, NJ). The test specimens were then scored independently by three investigators in a blinded fashion per following criteria: 0, no staining; 1+, minimal staining; 2+, moderate staining; 3+, strong staining. Areas of most intensity were identified by scanning tumor sections at low power (100×), and then images were photographed at high magnification (400×) using Olympus BX51 microscopy system with DP71 digital camera (Olympus, Melville, NY). Western blot analysis was performed per standard protocol.

Expression profiling of NSCLC-PDX and gene copy number analysis

Fresh HuPrime™ NSCLC tumor tissues were collected from the tumor-bearing mice, snap-frozen and stored at −80°C before being used for genetic and genomic analysis. For gene profiling analysis, the total RNA was isolated from the frozen tissues using Trizol (Invitrogen, Carlsbad, CA) per the manufacturer's instructions, and purified using RNeasy mini columns (Qiagen). RNA quality was assessed on a Bioanalyzer (Agilent). Only RNA samples with high quality (RIN>8) were used for expression profiling assays on Affymetrix HG-U219 array plates following standard protocol (http://media.affymetrix.com/support/downloads/manuals/3_ivt_express_kit_manual.pdf). Raw CEL data sets of all samples were normalized by RMA algorithm. Probe set intensity was expressed as log(2) transformed values. For SNP/CNV assay using Affymetrix SNP6.0 chips, genomic DNA was isolated and purified using Genomic DNA Tissue and Blood Isolation Kit (Qiagen) following manufacturer's instruction. DNA processing and chip hybridization were performed following standard Affymetrix protocol (http://media.affymetrix.com/support/downloads/manuals/genomewidesnp6_manual.pdf). Raw CEL data were QC-ed and filtered to remove low call-rate samples, and gene copy number analysis were performed by PICNIC and/or PennCNV methods. For some of the samples, the relative gene copy numbers were determined by qPCR. Briefly, the same genomic DNAs were subjected to amplification using MET specific primers (MET-F: GCTGGTGGTCCTACCATACATG; MET-R: CTGGCTTACAGCTAGTTTGCCA) by SYBR Green based quantitative PCR. The mammalian LINE-1 retrotransposon gene was used as a reference. The q-PCR data were analyzed on the chromo4 system using Opticon Monitor 3 software to generate the raw data. The raw data were then processed using the ΔCT relative quantification method. ΔCT= (CT value of target gene) − (CT value of reference gene). ΔCT values were then converted into intensity value (POWER(ΔCT-2)). All data were normalized to that of a sample with known MET copy number to obtain relative MET copy number.

Statistical analysis

The data of tumor volume were evaluated using Student's t-test for two comparisons, and one-way ANOVA test for multiple comparisons. All data were analyzed using SPSS 16.0. p < 0.05 was considered to be statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Identification of EGFR activating mutations in NSCLC HuPrime® models

We have established a large of panel of NSCLC PDX models by subcutaneous transplanting surgically removed tumor tissues from Asian patients in immunocompromised Balb/c nude mice. We performed hot-spot mutation analysis of these models on EGFR according to the known “classic” activating EGFR mutations (point/deletion mutations), frequently reported in the clinic.8 Specifically, HuPrime®-LU0858 carries L858R single point mutation; LU1868 contains L858R/T790M double mutations; LU1235 contains the common deletion mutation (exon 19: 2,236–2,350), and a number of models (e.g., LU1901 and LU2503) are with the wild-type EGFR. These model mutation status and their corresponding histopathology types are summarized in Tables 1, respectively. Furthermore, we also analyzed the available patient samples corresponding to the established models (LU2503 and LU1235) and confirmed the same EGFR status as in the models, suggesting that the models maintain most of the genetic defects as those in patients, similar to the earlier reports.19

Table 1. NSCLC HuPrime® models with different EGFR and c-met status
inline image

Interestingly, LU0858 with L858R and LU1868 with L858R/T790 point mutations both have SCC morphology as judged by several pathologists blindly and independently (Table 1, Supporting Information Table 1). This result is in contrast to that most of the reported mutations were exclusively found in ADC. Our interpretation for the difference is that there could be differences between Asian and Caucasian patients. In fact, there have been reports recently describing the EGFR point mutations in non-ADC (SCC, PLC and LCC) among Chinese26 and Japanese27 patients. It is worth mentioning that we also attempted to confirm the subtypes of these models by IHC analysis using differentiation markers (TTF-1, p63, CK-7 and CK5/6), but the results were not conclusive (data not shown).

Several other points are also worth mentioning. First, all the patients from whom the PDXs were derived had never been exposed to TKIs (naïve patients). This includes LU1868 with L858R/T790M double mutations, consistent with clinical observation.28–30 Second, the apparent higher EGFR mutations found in our NSCLC models likely reflect the known higher frequency of the mutations among Asian patient population, and the possible selection bias for mutated tumors during the engraftment process.

Treatment of HuPrime® LU1235 with activating EGFR deletion caused complete response to erlotinib

We were interested in investigating a variety of EGFR mutation status in the HuPrime® models in relation to their response to the EGFR inhibitors for predicting the response in patients. Prior to this, we set out to first confirm the above panel of models for expressing EGFR protein by IHC using monoclonal antibodies against total EGFR protein (tEGFR) and phosphorylated EGFR protein (activated, pEGFR). The result demonstrated positive staining of EGFR of both (Examples seen in Supporting Information Figure 1).

thumbnail image

Figure 1. Different NSCLC HuPrime® models respond to EGFR and cMET-inhibitors. (a) LU1235 complete responses to both erlotinib and cetuximab. (b, c) LU2503 and LU1901 effective responses to crizotinib at dosage of 50 mg/kg daily treatment respectively. (d) LU0858 responses to the treatments of vehicle, erlonitib, crizotinib and the combination of erlotinib and crizotinb at 50 mg/kg daily for 2 weeks. (e) LU1868 responses to erlonitib, crizotinib and cetuximab for 2–3 weeks. *p < 0.05; **p < 0.01 for significant difference when treatment group is compared to the vehicle control. The %ΔTC values were calculated for assessing tumor response to the treatment, where ΔT is tumor volume change in the treatment group and ΔC is tumor volume change in control group. (f) c-met gene copy and mRNA levels in different NSCLC HuPrime®. Upper panel: relative copy numbers determined by qPCR; Lower panel: relative mRNA levels determined by qRT-PCR (LUX and LUY are two NSCLC models without c-met gene amplification).

Download figure to PowerPoint

We then investigated drug activities by first examining whether LU1235 with the exon 19-deletion (2,236–2,350) would respond to erlotinib. Our results showed that daily 50 mg/kg dosing of erlotinib for 3 weeks caused complete remission (Fig. 1a). Tumors started to regress in the first week of treatment, and after 3 weeks of treatment, near-complete tumor response were observed with ΔTC = −7%. Our data suggest that the deletion mutation is indeed an oncogenic driver mutation for this particular tumor model, and the inhibition of EGFR suppressed the tumor development.

Next, we assessed whether LU1235 would also respond to cetuximab, a monoclonal antibody targeting EGFR. We thus tested cetuximab treatment (IP, 1 mg/mouse or ∼50 mg/kg, 3× weekly administrations), which also resulted in complete suppression of tumor growth (ΔTC = −12% at 3 week) (Fig. 1a). This suggests that cetuximab can effectively suppress EGFR activating mutation-driven oncogenesis in NSCLC, similarly as what is achieved with TKIs. This is consistent with the notion, suggested by others,31 that activation and high expression of EGFR is a positive factor that can contribute to the NSCLC response to cetuximab.

NSCLC HuPrime® models with amplified c-met gene, regardless of their EGFR status, respond to neither erlotinib nor cetuximab

We next investigated the HuPrime® models with activating L858R mutation (LU0858) and wild type (LU2503) for their responses to erlotinib. To our surprise, all these models, although naïve to TKIs, responded poorly or partially to erlotinib (Table 1, ΔTC = 107% LU2503, 59% for LU0858). These models (also LU1901, another model with wild-type EGFR) did not respond to cetuximab either (Table 1, ΔTC = 94% LU2503, 77% for LU1901 and 97% for LU0858). The non-/poor responsiveness to EGFR inhibitors suggested that there were at least additional oncogene driver(s) beyond EGFR activation for these tumors. Furthermore, these driver mutations, which cause the observed resistance, are pre-existing in patient tumors and did not result from selection pressures from EGFR inhibitors.

Since there were no obvious mutations in KRAS, BRAF, and no amplification in erbb2 and erbb3 genes, we investigated c-MET gene status. Interestingly, all these models (LU0858, LU1901 and LU2503) have c-MET gene amplification as demonstrated by qPCR analysis (Table 1 and Fig. 1F, upper panel). It is therefore plausible that c-MET could be candidate driver for the oncogenesis of these tumors, as reported in the clinic.12, 13 The c-met gene amplification was also found to be consistent with the increased expression of c-MET at mRNA levels as measured by Affymetric GenChip (Fig. 1F, lower panel), as well as the positive staining by IHC using monoclonal antibody against either total c-MET (t-MET) or activated c-MET (p-MET) (data not shown), in these models.

NSCLC HuPrime® with EGFR mutations and c-met amplification partially respond to crizotinib but completely respond to the combination of crizotinib and erlotinib

If c-met gene amplification plays role in the nonresponsiveness to EGFR inhibitors, these resistant models should respond to c-MET inhibitors. To this end, we first tested LU2503 and LU1901, two models with wild-type EGFR and with amplified c-met gene (Table 1), for response to crizotinib, the only marketed NSCLC target therapy with significant activity against c-MET.18 As anticipated, daily oral administration of crizotinib caused complete response of LU2503 and LU1901 (Fig. 1b, ΔTC = −11%, Fig. 1c, ΔTC = ∼8%, respectively, and Table 1), with tumor regression begun from day 3 post-treatment to the end of the study for LU2503. The reason for less effect for LU1901 than LU2503 is still unclear. In order to assess if there is a correlation between tumor response and inactivation of c-MET (dephosphorylation), we performed a single dose pharmacodynamic study on LU2053 that corresponds to the efficacy study shown in Figure 1d. The results demonstrated that crizotinib indeed inactivated p-c-MET as measured by Western blot analysis (Fig. 2a: p-c-MET vs. t-c-MET) and by IHC (Fig. 2b: p-c-MET), along with downstream biomarkers p-ERK and p-AKT (Fig. 2b). These results suggested that c-MET amplification is the main oncogenic driver for these two tumors. In order to rule out that the antitumor effect is not due to targeting EML4-ALK, another important activity, we confirmed that these two models (also LU0858 described below) do not have EML4-ALK fusion or other ALK mutation, and ALK expressions in these models are not elevated (data not shown). In addition, all these models responded to other highly specific investigational c-met inhibitors (Li, unpublished data).

thumbnail image

Figure 2. Pharmacodynamic analysis of LU2503 response to crizotinib. Single dose treatment with the same agents as described in Figure 1b. The tumor samples were harvested at the time points post-treatment as indicated in the figure. (a) Western blot analysis of different biomarkers of p-c-MET and total c-MET, along with GAPDH internal control. (b) IHC analysis (image) of different biomarkers of pERK images p-AKT and p-c-MET.

Download figure to PowerPoint

Next, we similarly evaluated LU0858 with L858R. Again, crizotinib caused significant but partial response (tumors stabilized at ∼200 mm3, ΔTC = 20%, Fig. 1d), although the effect was not as large as the complete response seen in the wild-type EGFR models of LU2503 and LU1901 (Figs. 1b and 1c). This observation suggests that the activated c-MET is one of the important oncogenic driver, but not as much as in LU2503 and LU1901. c-MET indeed played very important role in the resistance to the EGFR inhibitors. Since both crizotinib and erlotinib cause partial response in LU0858 (Fig. 1d), it is plausible that both c-MET and EGFR activation are the two key oncogenic drivers for this model. Indeed, when LU0858 was treated by a combination of erlotinib and crizotinib, the tumor completely disappear (ΔTC = −13%) (Fig. 1d), as seen in LU2503 with crizotinib monotherapy.

Furthermore, we also performed a similar single dose pharmacodynamic study on LU0858 (Fig. 3) that corresponds to the efficacy study shown in Figure 1d. The results demonstrated that crizotinib indeed inactivated pMET (Fig. 3e/3f) and downstream pERK (Fig. 3c/3d), as well as pEGFR (Fig. 3a/3b). It has recently been observed that c-MET can directly interact with and activate EGFR.13, 16 Our result pEGFR inactivation by crizotinib indeed supported this observation. Our results also showed that the inactivation of pEGFR, as well as the inactivation of downstream biomarker pERK, largely corresponds to the antitumor activity of the treatment.

thumbnail image

Figure 3. Pharmacodynamic analysis of LU0858 response to erlotinib and crizotinib. Single dose treatment with the same agents as described in Figure 1d. The tumor samples were harvested at the time points post-treatment as indicated in the figure (tarceva = erlotinib). IHC analysis of different biomarker: (a, b) pEGFR image and scores; (c, d) pERK images and scores; (e, f) p-c-MET image and scores. The scores were the averages of readings from three independent and blinded researchers.

Download figure to PowerPoint

HuPrime® LU1868 with L858R/T790M without c-met amplification does not respond to erlotinib or crizotinib, but highly respond to cetuximab

LU1868, also from naïve patient, was found containing L858R/T790M double mutations. This type of mutation is most frequently associated with the resistance after TKI exposure in the clinic (∼50%), but usually does not overlap with c-met amplification in the same patients.9 When LU1868 was treated with erlotinib and crizotinib, the tumor responded to neither as expected (Fig. 1e) (ΔTC = 84% and 132%, respectively), confirming those seen in the clinic. Since we observed the sensitivity of LU1235 (with deletion in EGFR) to cetuximab, we thus tested whether LU1868 responded to cetuximab. Similarly to LU1235 (Fig. 1a), LU1868 is highly responsive to cetuximab (Fig. 1e, ΔTC = −6%). This is also consistent with previous reports that these mutations are positive factors for the cetuximab response.31 However, our data using PDX models showing that complete response can be achieved to cetuximab, in contrast to only partial response was achieved in mouse NSCLC tumor model of transgenic EGFR T790M.32 Nevertheless, this observation suggests that cetuximab could potentially be an effective target treatment for the double mutation containing TKI-resistant patients, if confirmed in the clinic.

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Erlotinib and gefitinib, either as monotherapy or in combination with chemotherapy, have shown limited overall benefit for NSCLC patients, largely due to two reasons: first, only small percentage of ADC patients with EGFR activating mutations would likely be benefited; and second, treatment usually result in development of drug resistances with two dominant distinct mechanisms. Moreover, both these resistances can also be detected in naïve patient populations without prior exposure to TKIs as demonstrated by the present study, as well as reports by others.13 It is apparently of great medical importance to identify approaches to overcome patients' resistances.9 Several second-generation investigational irreversible inhibitors are being developed to overcome the T790M induced resistance, but yet to be approved. A number of c-MET inhibitors are also being developed in the clinic but has yet to be approved too.

Crizotinib, a recently approved target treatment only for the NSCLC with ALK-EML4 fusion (only 3–5% of NSCLCs), was also a c-MET inhibitor.18 The present study using PDX for the first time demonstrated that crizotinib, particularly in combination with TKI, could potentially be used to overcome c-MET amplification caused TKI resistance in NSCLC patients (Table 1), consistent with previous reports that the sensitivity to TKIs can be restored by c-met inhibitor in the c-MET activation induced TKI-resistant cells, an observation made based cell line.13 Our work here also suggests for the first time that the resistance caused by L858R/T790M double mutations can be overcome by cetuximab. In particular, together with the complete response to cetuximab seen in LU1235, we could also potentially predict that many NSCLC patients with the EGFR activating mutations but without c-met gene amplification would likely respond to cetuximab, and/or other TKIs, particularly those overcoming the T790M mediated resistance.

Although crizotinib and cetuximab have yet to be approved for the exact indications as proposed here (c-met amplified NSCLC and EGFR mutated NSCLC without c-met amplifications, respectively), our suggested utilities of these two drugs should be readily confirmed in the clinic due to the availability of the clinically ready companion diagnosis. These diagnoses include the approved EGFR mutation test and FISH assay for c-MET gene amplification.

PDX models are believed to have similar histopathological and genetic profiles as those of the patients' from whom they are derived19–24 with arguably the best predictive power to forecast clinical outcomes. Confirmation of our observations in the clinic would not only provide patients with new and effective treatments but would also further validate this predictive power.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors thank the patients who donated their tissues for this study. They thank scientists from BMS who have profiled some of these HuPrime models® and help to understand genomic/genetic background of these models (Drs. Heshani DeSilva, Petra Ross-Macdonald, Aiqing He, Xiadi Zhou, Matthew Lorenzi, Marco Gottardis and Rolf-Peter Ryseck). They also thank the technicians at Division of Translational Oncology and Animal Center at Crown Bioscience for technical support of this work.

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_27813_sm_SuppFig1.tif4076KSupporting Information Figure 1
IJC_27813_sm_SuppFig2A.tif1164KSupporting Information Figure 2A
IJC_27813_sm_SuppFig2B.tif1995KSupporting Information Figure 2B
IJC_27813_sm_SuppFig2C.tif2111KSupporting Information Figure 2C
IJC_27813_sm_SuppFig2D.tif2003KSupporting Information Figure 2D
IJC_27813_sm_SuppFig2E.tif2064KSupporting Information Figure 2E
IJC_27813_sm_SuppTab1.pdf81KSupporting Information Table 1

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.