High-MET status in non-small cell lung tumors ...

(This is a sample cover image for this issue. The actual cover is not yet available at this time.)

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the author's institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier's archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights

Author's Personal Copy Lung Cancer 101 (2016) 59–67

Contents lists available at ScienceDirect

Lung Cancer journal homepage: www.elsevier.com/locate/lungcan

High-MET status in non-small cell lung tumors correlates with receptor phosphorylation but not with the serum level of soluble form Marie-Christine Copin a,b , Marie Lesaffre a , Mélanie Berbon a , Louis Doublet d , Catherine Leroy a , Emmanuelle Tresch d , Henri Porte e , Jérôme Vicogne a , Alexis B. Cortot a,c , Eric Dansin d , David Tulasne a,∗ a

Univ. Lille, CNRS, Institut Pasteur de Lille, UMR 8161 – M3T – Mechanisms of Tumorigenesis and Targeted Therapies, F-59000 Lille, France Univ. Lille, Institut de Pathologie, CHU Lille, Avenue Oscar Lambret, F-59000 Lille, France c Univ. Lille, CHU Lille, Thoracic Oncology Department, F-59000 Lille, France d Département de Cancérologie Générale, CLCC Oscar Lambret, 3 rue Fréderic Combemale, Lille 59020, France e Service de Chirurgie Thoracique, Centre Hospitalier Régional Universitaire de Lille, Lille, France b

a r t i c l e

i n f o

Article history: Received 24 June 2016 Received in revised form 1 September 2016 Accepted 13 September 2016 Keywords: Lung cancer MET Receptor tyrosine kinase Hepatocyte growth factor/scatter factor Phosphorylation Proteolytic cleavages

a b s t r a c t Objectives: The receptor tyrosine kinase MET is essential to embryonic development and organ regeneration. Its deregulation is associated with tumorigenesis. While MET gene amplification and mutations leading to MET self-activation concern only a few patients, a high MET level has been found in about half of the non-small cell lung cancers (NSCLCs) tested. How this affects MET activation in tumors is unclear. Also uncertain is the prognostic value, in cancer, of a phenomenon well described in cell models: MET shedding, i.e. its cleavage by membrane proteases leading to release of a soluble fragment into the medium. Materials and methods: A prospective cohort of 39 NSCLC patients was constituted at diagnosis or soon after. Normal tissues, tumor tissues, and blood samples were obtained. This allowed, for the same patient, synchronous determination of (i) the MET level in the tumor, (ii) receptor phosphorylation, and (iii) the concentration of soluble MET fragment (sMET) in the serum. Results: After confirming the adequacy of an ELISA for measuring the serum level of sMET, we found no correlation between this level and the concentration of MET in tumors, as evaluated by immunohistochemistry and western blotting. Nevertheless, all but one tumor displaying a high MET level also displayed receptor phosphorylation, restricted to a small number of tumor cells. Conclusion: Our results thus demonstrate that the serum level of sMET is not indicative of the amount of MET present in the tumor cells and cannot be used as a biomarker for therapeutic purposes. However, MET scoring of tumor biopsies could be a first step prior to determination of MET receptor activation in high-MET tumors. © 2016 Published by Elsevier Ireland Ltd.

1. Introduction MET is a receptor tyrosine kinase (RTK) present predominantly in cells of epithelial origin, activated by its stromal ligand, the hepatocyte growth factor/scatter factor (HGF/SF) [1]. HGF and MET are essential to embryonic development, since knockout of either one affects epithelial organ, placenta, muscle, and neuron formation [2–5]. Conditional knockout of MET in the lung inhibits alveolar

∗ Corresponding author. E-mail address: [email protected] (D. Tulasne). http://dx.doi.org/10.1016/j.lungcan.2016.09.009 0169-5002/© 2016 Published by Elsevier Ireland Ltd.

development, possibly because of decreased alveolar epithelial cell proliferation and survival [6]. Aberrant MET and HGF signaling is involved in promoting tumorigenesis and metastasis [7]. MET is overproduced in half of all non-small-cell lung cancers (NSCLCs) which is associated with poor prognosis [8,9]. Furthermore, MET mutations have been discovered in a variety of cancers. In renal cancers, most are located in the kinase domain and cause kinase activition [10]. Recently discovered in lung cancers are various mutations, leading to exon skipping and to deletion in the regulatory juxtamembrane domain [11,12]. Furthermore, in about 5–20% of NSCLCs displaying a mutated Epidermal Growth Factor Receptor (EGFR) gene, acquired resistance to EGFR inhibitors involves amplification of the MET gene,

Author's Personal Copy 60

M.-C. Copin et al. / Lung Cancer 101 (2016) 59–67

associated with MET self-activation [13]. In recent phase III clinical trials evaluating the anticancer effect of anti-MET antibodies or tyrosine kinase inhibitors (TKIs) in patients stratified according to the amount of MET detected by immunohistochemistry (IHC), these agents have proved ineffective. Yet several studies have revealed encouraging responses in a limited number of patients displaying MET gene amplification or mutations [12,14–18]. Upon ligand binding and subsequent dimerization, MET autophosphorylates tyrosine residues located notably in its kinase domains [19,20]. MET phosphorylation has been detected in tumor samples, including NSCLC samples, but whether the MET level correlates with MET phosphorylation remains controversial [21,22]. Downregulation of the MET receptor is an essential desensitization mechanism. In addition to the well-known ligand-dependent degradation of MET, we have demonstrated that MET undergoes several proteolytic cleavages promoting its downregulation [23–25]. For instance, under steady-state condition, MET is processed by Presenilin-Regulated Intramembrane Proteolysis (PSRIP). This proteolytic process involves a first cleavage by ADAM metalloproteases, leading to generation of an N-terminal fragment (soluble MET; sMET), which is released into the extracellular medium, and a membrane-anchored p55MET fragment, which is degraded by the lysosome [26], or further cleaved by ␥-secretase, yielding an intracellular p50MET fragment degraded by the proteasome [27]. These cleavages help reduce the half-life of the receptor. HER2 likewise undergoes proteolytic cleavages by membrane metalloproteases, leading to generation of a soluble HER2. Interestingly, this soluble fragment is detectable in the sera of patients with breast cancer, where its level correlates with the level of full-length HER2. It is thus a potential serum biomarker of HER2 accumulation [28,29]. We thus wondered whether the sMET fragment might provide a relevant biomarker of MET accumulation. To answer this question, we have characterized in the same NSCLC patients the level of MET receptor in paraffin-embedded sections and frozen tumors, its phosphorylation state, and the serum concentration of sMET fragment.

(6C5-32233) was purchased from Santa Cruz Biotechnology. Rabbit monoclonal antibody directed against the C-terminal tail of MET (SP44) used for immunohistochemistry (IHC) was purchased from Roche (Schlieren, Switzerland). HRP-conjugated monoclonal antibody directed against the histidine tag was purchased from Invitrogen. Mouse monoclonal antibody directed against the Nterminal domain of MET (Met extra) (DL-21) was kindly provided by Dr. Silvia Giordano (University of Torino Medical School, Italy). 2.3. Tumor sample preparation Tumor samples were divided into two parts; frozen in a Snapfrost fast freezing system (Excilone, Aperio, CA, USA) and stored at −80 ◦ C; formalin-fixed and paraffin-embedded (FFPE). For western blot analysis, frozen samples were sliced into pieces about 1.2 mm in diameter and transferred into Lysing Matrix type D tubes containing ceramic beads (MP Biomedicals, Santa Ana, CA, USA) in the presence of RIPA buffer (50 mM Tris-HCl pH = 7.4, 150 mM NaCl, 2 mM EDTA, 50 mM NaF, 0.5% sodium deoxycholate, 0.1% SDS, and 1% NP-40). The samples were lysed with a FastPrep homogenizer (MP Biomedicals) (4 cycles of 40 s at 6 m/s, each followed by a 5 min pause on ice). The samples were then centrifuged and proteins in the supernatant were quantified by the BCA Protein Assay (Pierce, Rockford, IL, USA). Protein were analyzed by western blotting. For IHC, FFPE tissue sections were stained with Hematein/Eosin/Safran and IHC was performed in an automated immunostainer Benchmark XT (Ventana Medical Systems) with an antibody against the intracellular domain of MET (SP44 CONFIRM, Ventana Medical Systems). MET was scored according to the study of Spigel et al. [8] (score 3: high-intensity staining of at least 50% of the tumor; score 2: moderate staining of at least 50% of the tumor and high-intensity staining of less than 50%; score 1: weak staining of at least 50% of the tumor and moderate or strong staining of less than 50%; score 0: no staining or staining at any intensity of less than 50% of the tumor). 2.4. Western blotting

2. Materials and methods Western blotting was performed as previously described [30]. 2.1. Cell cultures 2.5. RNA interference The breast cancer cell line MDA-MB-231 and the gastric cancer cell line GTL-16 were cultured in Dulbecco’s Modified Eagle’s Medium (Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS, Life Technologies) and antibiotics. The lung cancer cell line EBC-1 was cultured in EMEM medium supplemented with 10% FBS and antibiotics. MCF-10A mammary epithelial cells were cultured in DMEM and HAM’s F12 vol/vol (Life Technologies) supplemented with 5% horse serum (Life Technologies), 500 ng/ml hydrocortisone (Calbiochem), 20 ng/ml epidermal growth factor (Peprotech), 10 ␮g/ml insulin (Sigma Aldrich), 100 ng/ml cholera toxin (Calbiochem), and 1% antibiotics. Primary human endothelial cells (HUVECs) were cultured in EGM2 medium (Lonza). Primary human keratinocytes (NHEKs) were cultured in KGM gold keratinocyte growth medium (00192060, Lonza). The lung cell line p339 (6CFSMEO-) was cultured in MEM, 10% FCS, antibiotics. 2.2. Antibodies Rabbit polyclonal antibodies directed against phosphorylated (Y1234/1235) MET (#3126) were purchased from Cell Signaling Technology (Danvers, MA, USA). Mouse monoclonal antibody against the kinase domain of MET (3D4) was purchased from Life Technologies. Mouse monoclonal antibody against GAPDH

A suspension of 400,000 cells was incubated for 20 min with a mix of 4.5 ␮l/ml Lipofectamine 2000 (Invitrogen) and 3 nM (MCF10A cells) or 120 nM (GTL16 cells) siRNA. The cells were then plated in complete medium in a 6-well plate. The MET-targeting siRNAs were a pool of three stealth siRNAs (Invitrogen) [5 -CCAUUUCAACUGAGUUUGCUGUUAA3 , 5 -UCCAGAAGAUCAGUUUCCUAAUUCA-3 , 5 CCGAGGGAAUCAUCAUGAAAGAUUU-3 ]. A negative control Stealth siRNA was also purchased from Invitrogen. 2.6. Enzyme-linked immunosorbent assay The concentration of sMET in conditioned medium was measured by an ELISA performed with the c-MET (soluble) ELISA Kit, human (Novex KHO2031). Briefly, patient serum was diluted 1:100 and conditioned serum-free medium from cell lines was diluted 1:100 for EBC-1 and GTL-16 cells or 1:3 for the other cells. One hundred microliters of the dilution was incubated for 2 h in a plate coated with antibodies against human soluble MET. Detection was done with an anti-human soluble MET antibody-biotin/streptavidin-HRP. After incubation for 30 min with chromogenic substrate, the plate was read at 450 nm with a Multiskan RC (Thermo Labsystem) spectrophotometer.

Author's Personal Copy M.-C. Copin et al. / Lung Cancer 101 (2016) 59–67 Table 1 Clinical data pertaining to the 39 NSCLC patients. Characteristics (N = 39)

n

%

Sex Men Women

27 12

69.2% 30.8%

Age (Year) N = 39 Median – (extremes) Average – standart deviation

61 60.2

(17–86) 12.4

Histologic type Adenocarcinoma Squamous cell carcinoma Large cell carcinoma Sarcomatoid carcinoma

17 15 6 1

43.6% 38.5% 15.4% 2.6%

T Stage (tumor size) T1 T1A T1B T2 T2B T3 T4

5 4 10 2 3 3 12

12.8% 10.3% 25.6% 5.1% 7.7% 7.7% 30.8%

N Stage (lymph node) N0 N1 N2 N3 NX

13 6 14 5 1

33.3% 15.4% 35.9% 12.8% 2.6%

M Stage (metastasis) M0 M1 M1B

20 12 7

51.3% 30.8% 17.9%

Global stage I II III-A III-B IV

9 3 5 3 19

23% 8% 13% 8% 48%

Previous Neoadjuvant chemotherapy Yes No

2 37

5.1% 94.9

ECOG Performance Status 0 1 2 3

20 16 2 1

51.3% 41.0% 5.1% 2.6%

Smoking Former smoker Smoker Non-smoker

28 8 3

71.8% 20.5% 7.7%

61

C-terminal tail of the receptor. The MET score was determined according to the study of Spigel et al. [8]. Four patients displayed score 0, twelve score 1, sixteen score 2, and seven score 3 (Table 2). Staining patterns representative of these scores are shown in Fig. 1. Consistently with previous studies, about half of the patients displayed MET overexpression (score 2 or 3) [8,9]. All normal tissues displayed weak staining corresponding to MET score 1. 3.2. Validation of sMET quantification by ELISA To validate the ELISA used to quantify the sMET fragment, we compared the results obtained when the ELISA on the one hand, and western blotting on the other were applied to cell lines characterized by different MET levels. In gastric GTL16 cells and pulmonary EBC-1 cells displaying MET gene amplification, western blot analysis confirmed high levels of full-length MET in cell lysates, and a band at about 90 kDa, corresponding to sMET, was easily detected in the medium. In the mammary cell lines MCF10A and MDA-MB231, where the MET level was lower, the amount of sMET detected was also lower (Fig. 2A). Consistently, the ELISA revealed 1.4 ␮g/ml sMET in the GTL16- and EBC-1-cell-conditioned media, but no more than 0.1 ␮g/ml in the other cell-conditioned media (Fig. 2A). MET silencing in GTL16 and MCF10A cells led to a drastic decrease in sMET production, as confirmed by both western blotting and the ELISA (Fig. 2B). In crude serum, sMET was not detectable by western blotting, probably because of the high concentration of proteins. Therefore, we used biotinylated K1 protein (a high-affinity specific ligand of MET) to capture sMET [26,31]. This specific pull-down allowed easier detection of the fragment by western blotting (Fig. 2C) [26]. Next, we used a recombinant extracellular domain of human MET (residues 23 to 932) fused to the human IgG Fc domain followed by a 6× histidine tag as positive control for detection. To test the ability of the ELISA to measure sMET in serum, increasing concentrations of recombinant sMET chimera were added to FBS. MET-Fc was detected in a dose-dependent manner in the FBS (Fig. 2D). It is worth noting that bovine sMET was not recognized by the anti-human MET antibodies in this ELISA. To further evidence the presence of sMET in serum, a fixed concentration of recombinant MET-Fc was incubated with increasing concentrations of human serum and recombinant MET-Fc was specifically detected with an anti-histidine antibody. The signal due to recombinant MET-Fc decreased as the concentration of human serum increased, suggesting competition between sMET of serum and MET-Fc (Fig. 2E). Taken together, our data demonstrate that the ELISA used allows quantitative detection of sMET in serum. 3.3. Serum levels of sMET do not correlate with tumor MET scores

3. Results 3.1. Paraffin-embedded, frozen tumors and blood samples from a prospective NSCLC cohort The prospective cohort of 39 NSCLC patients included 17 patients with adenocarcinoma, 15 with squamous cell carcinoma, 6 with large cell carcinoma, and 1 with a sarcomatoid carcinoma (Table 1). Biopsies were obtained at diagnosis from 26 naïveof-treatment patients with advanced or metastatic disease and surgical samples were obtained from 13 patients who had undergone surgical resection. Among these 13 patients, 11 were naïve of treatment and two had been treated by neoadjuvant chemotherapy (taxotere/cisplatin) before surgery. Tumors and adjacent normal tissues were collected and frozen for western blotting or formalin fixed and paraffin embedded for IHC. The MET level was evaluated by IHC performed on paraffinembedded tissue sections with an antibody targeting the

The sMET concentration was measured by ELISA in the sera from the 39 NSCLC patients and 11 healthy volunteers. The concentration ranged from 368 to 1115 ng/ml in patients (except one at 92 ng/ml), with an average of 732 ng/ml (Fig. 3A). When the patients were divided into a “high MET” and a “low MET” group according to their MET IHC scores, and regardless of whether the “high MET” group included scores 2 and 3 or only score 3, no significant difference in serum sMET levels was observed between the two groups (Fig. 3B). Since both healthy volunteers and NSCLC patients showed rather high sMET serum levels, we evaluated sMET production in various primary, immortalized, and cancer cell lines including HUVECs, NHEKs, the immortalized mammary cell line MCF10A, the immortalized lung cell line p339 (6CFSMEO-), and the gastric carcinoma cell line GTL16. All these cell types were found to release significant concentrations of sMET into the medium, even when full-length MET was barely detectable in cell lysates as for HUVECs (Fig. 3C). These data suggest that the high level of sMET

Author's Personal Copy 62

M.-C. Copin et al. / Lung Cancer 101 (2016) 59–67

Table 2 Histologic type, MET score, and MET phosphorylation status of each NSCLC patient. Patient

Histologic type

MET score

Phospho-MET Status

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

Adenocarcinoma Squamous cell carcinoma Adenocarcinoma Non-small cell carcinoma Squamous cell carcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Adenocarcinoma Large cell neuroendocrine carcinoma Adenocarcinoma Adenocarcinoma Squamous cell carcinoma Large cell neuroendocrine carcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Large cell carcinoma Adenocarcinoma Adenocarcinoma Squamous cell carcinoma Non-small cell carcinoma Adenocarcinoma Squamous cell carcinoma Squamous cell carcinoma Adenocarcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Adenocarcinoma Adenocarcinoma Squamous cell carcinoma Adenocarcinoma Non-small cell carcinoma Adenocarcinoma Squamous cell carcinoma

3 1 1 2 1 0 2 3 2 2 1 2 0 2 0 1 2 2 3 1 0 2 3 2 1 3 2 1 3 2 2 1 2 2 1 3 1 2 1

Positive 0 0 0 0 0 0 Positive 0 0 0 0 0 0 0 0 0 0 Positive 0 0 0 Positive 0 0 0 0 0 Positive 0 0 0 0 0 0 Positive 0 0 0

Histologic type was determined after Hematein/Eosin/Safran staining. MET was detected on FFPE samples by IHC with an antibody directed against the C-terminal tail of the receptor. A MET score (0–3) was determined according to Spigel et al. [8]. Phosphorylated MET was detected with an antibody directed against the phospho-tyrosine residues 1234 and 1235. A sample was considered positive when at least 5% of the tumor cells displayed staining.

Fig. 1. Representative picture for each MET score determined by IHC. MET was detected on FFPE samples by IHC with an antibody directed against the MET C-terminal tail. A picture representative of each of the four MET scores (0, 1, 2 and 3) is shown (patients 6, 16, 34, and 1, respectively).

Author's Personal Copy M.-C. Copin et al. / Lung Cancer 101 (2016) 59–67

63

Fig. 2. Validation of the ELISA used to detect sMET in serum. (A) EBC-1, GTL-16, MCF-10A, and MDAMB-231 cells were cultured in serum-free medium. The conditioned medium was collected and concentrated. Whole-cell lysates were prepared with the corresponding cells. (B) GTL-16 and MCF-10A cells were transfected with scrambled or targeting MET siRNA. The following day, the conditioned serum-free media were collected and concentrated and the corresponding whole-cell lysates were prepared. (A, B) Equal amounts of protein were resolved by 10% SDS PAGE and analyzed by western blotting with antibodies directed against the luminal region of the MET ␤-chain (DL21) and against GAPDH to assess loading. The positions of prestained molecular weight markers are indicated. Arrows indicate the positions of precursor and mature full-length MET and sMET. An ELISA was used to quantify sMET in the conditioned serum-free media obtained after culturing EBC-1, GTL-16, MCF-10A, and MDA-MB-231 cells (n = 3; −/+ SD). (C) Serum-free medium conditioned by GTL-16, serum from a healthy person, and non-conditioned medium containing or not the recombinant MET extracellular region fused to the Fc domain (MET-Fc) were collected and exposed to capture by K1 (binding domain of HGF). Proteins in the captured fractions were resolved by 10% SDS PAGE and analyzed by western blotting with antibodies directed against the luminal region of the MET ␤-chain (DL21). The positions of prestained molecular weight markers are indicated. Arrows indicate the positions of recombinant MET-Fc and sMET. (D) Increasing concentrations of recombinant extracellular domain of human MET were added to fetal bovine serum (FBS) and quantified by ELISA. (E) 50 ng/ml MET-Fc was incubated with increasing concentrations of human serum. MET-Fc was specifically detected by means of the modified ELISA using an anti-histidine antibody as secondary antibody.

detected in serum could be due to massive physiological MET shedding, including from endothelial cells directly in contact with the bloodstream. 3.4. The highest MET levels correlate MET tyrosine phosphorylation The MET phosphorylation status was evaluated on tumor sections by IHC with an antibody directed against the phosphorylated

tyrosine residues 1234 and 1235 of the kinase domain. None of the 32 patients with MET scores 0, 1, or 2 displayed MET phosphorylation, while 6 of the 7 patients with score 3 were positive (Table 2). Thus, in this cohort, MET phosphorylation is significantly associated with the highest MET score (Fig. 4A). MET phosphorylation was restricted to a small proportion (5–20%) of the tumor cells, located preferentially at the invasion front or in clusters of tumor cells spreading through alveolar spaces (Fig. 4B) (Copin et al. Manuscript in preparation).

Author's Personal Copy 64

M.-C. Copin et al. / Lung Cancer 101 (2016) 59–67

Fig. 3. Measure of sMET level in NSCLC. (A) Sera from 11 healthy persons and 39 NSCLC patients were analyzed by ELISA to determine the concentration of sMET. The 39 NSCLC sera were classified according to their total IHC-based MET score. (B) Median and average sMET concentrations were measured by ELISA in two groups, defined as “high MET” and “low MET” according to their MET scores determined by IHC. High MET group included either scores 2 and 3 or only score 3. Statistical differences in sMET concentration between the two groups were assessed by Student t-test. (C) Serum-free media conditioned by GTL-16, MCF-10A, HUVEC, NHEK, and P339 cells were collected and concentrated. Whole-cell lysates were prepared from the corresponding cell cultures. Equal amounts of protein were resolved by 10% SDS PAGE and analyzed by western blotting with antibodies directed against the luminal region of the MET ␤-chain (DL21) and against GAPDH to assess loading. The positions of prestained molecular weight markers are indicated. Arrows indicate the positions of precursor and mature full-length MET and sMET.

To assess MET phosphorylation, cell lysates from the 13 surgical samples were analyzed by western blotting with anti-MET or anti-phospho-MET antibodies. In all normal lung tissues, MET was barely detectable and its phosphorylated form was undetectable (Fig. 4C). In contrast, tumor tissues showed overall a MET signal intensity in agreement with the score determined by IHC. Phosphorylated MET was detectable only in the two score-3 tumors and one score-2 tumor displaying high MET expression level (Fig. 4C). Thus, phosphorylated receptors are present in samples having the highest MET levels.

4. Discussion Many MET-targeting treatments are evaluated in clinical trials [7]. However, an important issue is how to select those patients who are liable to benefit from these treatments. Recent phase III clinical trials on NSCLC patients found treatment with a MET tyrosine kinase inhibitor or with a MET-blocking monoclonal antibody combined with an EGFR tyrosine kinase inhibitor to be ineffective. This failure may be due to inadequate patient stratification, based solely on the MET level determined by IHC. Here we have evaluated in each patient, at the time of diagnosis or soon after, both the

Author's Personal Copy M.-C. Copin et al. / Lung Cancer 101 (2016) 59–67

65

Fig. 4. Phosphorylation of MET tyrosines as assessed by IHC and western blotting. (A, B) MET and phosphorylated MET were detected on FFPE samples by IHC with an antibody directed against the C-terminal tail of MET or against its phosphorylated tyrosine residues 1234 and 1235. (A) The 39 patients were divided in two groups, one comprising the patients with MET scores 0, 1, and 2 (31 patients) and one comprising the score-3 patients (7 patients). In the first group, no patient was positive for phospho-MET, while in the second one 6 out of 7 patients displayed MET phosphorylation. Detection of phosphorylated MET was significantly associated with score 3 (p < 0.001 Fisher’s exact test). (B) Pictures of representative of samples displaying MET score 3 and receptor phosphorylation are shown (patients 8, 23, and 36). (C) For the 13 tumor samples (t = tumor) and their corresponding adjacent normal tissues (n = normal) collected by surgery, equal amounts of protein were resolved by 10% SDS PAGE and analyzed by western blotting with antibodies directed against the MET kinase domain, phosphorylated MET (phospho-MET), and GAPDH (to assess loading). The positions of prestained molecular weight markers are indicated. Arrows indicate the positions of precursor and mature full-length MET.

intensity of MET staining and the level of MET phosphorylation in the tumor, along with the level of sMET in the serum. Like other receptor tyrosine kinases, MET undergoes cleavages by metalloproteases, leading to generation of an extracellular fragment. We have previously demonstrated that MET shedding is a constitutive process, with MET cleavage being proportional to the level of full-length receptor [27]. Accordingly, in mice xenografted with MET-overexpressing cells, tumor size was found to correlate with the sMET in the serum [40]. Likewise, the amount of full-length HER2 detected in a breast tumor correlates with level of soluble fragment in the patient’s serum. In the present study, strikingly, we observe no correlation between the level of sMET in the serum and the amount of

full-length MET in the tumor. Also interesting are the rather high sMET levels observed in sera from healthy volunteers, which is not significantly different from the average for patients. Both of these findings might reflect the fact that many cell types naturally produce high amounts of sMET liable to contribute to the serum level. In particular, a major contribution might be made by endothelial cells. Any additional generation of sMET by a MET-overexpressing tumor might be insufficient to significantly increase the level of sMET in the blood. Recently, soluble MET expression in serum of lung cancer patient was evaluated in two Chinese cohorts. In these studies the sMET level was found to correlate with the highest level of full-length MET within the tumor [32,33]. Detection methods of

Author's Personal Copy 66

M.-C. Copin et al. / Lung Cancer 101 (2016) 59–67

sMET by ELISA were similar between our and these studies allowing reasonable comparison of the data. In healthy volunteers and in patients who did not overexpress MET, sMET was detected at about 700–1000 ng/ml, consistent with the level found here. In patients overexpressing MET, however, the sMET concentration reached 1900 ng/ml, a level never reached in our study. However, constitutions of the cohorts were quite different between the studies. First, our cohort was prospective with a recruitment of the patients at the diagnosis or soon after. Consequently, only two patients among 39 received adjuvant chemotherapy before surgery. In addition, in our cohort, EGFR mutations were absent in the patients tested, accordingly to low rate of EGFR mutation in European population. In contrast, the two Chinese cohorts were retrospective and patients display high rate of EGFR mutations accordingly to the rate found in the Asiatic population. In the study of Fu et al., all the patients selected harbored EGFR mutations and were all treated with an EGFR inhibitor [32]. In the study of Lv et al., about a quarter of the NSCLC patients displayed EGFR mutations and received erlotinib [33]. Treatments with other chemotherapies were not documented. Importantly, treatment is well known to influence the molecular profile of the lung tumors. For instance, resistance to EGFR-targeting therapies has been linked to amplification of the MET gene, leading to its massive expression in about 20% of patients [13]. Thus, the very high level of sMET in the Chinese cohort could be due to MET gene amplification. The MET gene amplifications were not documented in these studies. Furthermore, the patients showing the highest sMET levels in the Fu et al. study had tumors diameters exceeding 7 cm [32], a tumor size not reached in our cohort. Taken together, these suggest that sMET can exceed the basal level of about 1000 ng/ml only in the case of large tumors overexpressing MET, possibly as a result of the MET gene amplification observed in resistant patients. Using an antibody directed against phosphorylated Y12341235, indicative of MET activation, we found constitutive MET phosphorylation in only six tumors: all but one of the seven tumors with MET score 3. We recently confirmed the association between a high MET score and phosphorylation in an independent retrospective cohort, since phospho-MET was detected in 5–20% of the tumor cells in 10 of the 33 adenocarcinomas, with a significant association between MET phosphorylation and a high MET score (Lapère et al., submitted manuscript). Interestingly, several studies performed on NSCLC cohorts have found no association between a high MET level and MET phosphorylation [22], while other studies have [21]. It is worth noting that although MET was detected in most of the cells of the score 3 tumors, its phosphorylation was detected only in a restricted contingent of cells (about 5–20%). Interestingly, some studies have been performed on tissue microarrays (TMAs) covering a limited area of the tumors [22]. A TMA might not be representative of the whole tumor and might miss clusters of cells displaying MET phosphorylation. In several type of cancer, detection of phosphorylated MET was associated with poor prognosis highlighted interest of its efficient detection. This was true for instance for head and neck, gastric or breast cancers [34–36]. In small cell lung cancer (SCLC), MET activation was also associated with worse prognosis [37]. The situation is less clear for NSCLC, since several studies failed to associate MET phosphorylation to poor prognosis [38,39]. However, for a restricted number of NSCLC patients, MET activation through either MET gene amplification or activating mutations lead to clear ligand independent MET activation. In addition, several case reports described response to treatment with a TKI against MET for the patients harboring such MET genetic alterations [12,14–18]. Nevertheless, each of these genetic anomalies is found in less than 3% of all NSCLCs. Importantly, we show here that patients with the highest MET score (score 3) displayed MET receptor

phosphorylation, suggestive of receptor activation. Since patients displaying MET phosphorylation represent more than 15% of NSCLC cases, it could be interesting to evaluate their response to MET TKIs. In conclusion, the serum level of sMET fragment measured at diagnosis is not an adequate biomarker of full-length MET accumulation in the tumor. In our cohort, however, nearly all the high-MET tumors showed MET receptor activation, liable to make them responsive to treatment with MET inhibitors. Conflicts of interest The authors declare that there are no conflicts of interest. Acknowledgments This work was supported by the CNRS, the “Institut Pasteur de Lille”, and INSERM, and by grants from the “Ligue Contre le Cancer, Comité Nord”, the “Association pour la Recherche sur le Cancer”, the “Institut National du Cancer”, the “Cancéropôle Nord-Ouest”, the “Site de Recherche Intégrée sur le Cancer, SIRIC ONCOLille”, and the Roche company. We thank the Microscopy-Imaging-Cytometry Facility of the BioImaging Center Lille Nord-de-France for access to instruments and technical advice and the Chemistry Systems Biology (CSB) facility for its technical advice and support. References [1] C. Birchmeier, W. Birchmeier, E. Gherardi, G.F. Vande Woude, Met, metastasis, motility and more, Nat. Rev. Mol. Cell Biol. 4 (2003) 915–925. [2] F. Bladt, D. Riethmacher, S. Isenmann, A. Aguzzi, C. Birchmeier, Essential role for the c-met receptor in the migration of myogenic precursor cells into the limb bud, Nature 376 (1995) 768–771. [3] C. Schmidt, F. Bladt, S. Goedecke, V. Brinkmann, W. Zschiesche, M. Sharpe, E. Gherardi, C. Birchmeier, Scatter factor/hepatocyte growth factor is essential for liver development, Nature 373 (1995) 699–702. [4] Y. Uehara, O. Minowa, C. Mori, K. Shlota, J. Kuno, T. Noda, N. Kitamura, Placental defect and embryonic lethality in mice lacking hepatocyte growth factor/scatter factor, Nature 373 (1995) 702–705. [5] F. Maina, M.C. Hilton, C. Ponzetto, A.M. Davies, R. Klein, Met receptor signaling is required for sensory nerve development and HGF promotes axonal growth and survival of sensory neurons, Genes Dev. 11 (1997) 3341–3350. [6] C. Calvi, M. Podowski, A. Lopez-Mercado, S. Metzger, K. Misono, A. Malinina, D. Dikeman, H. Poonyagariyon, L. Ynalvez, R. Derakhshandeh, A. Le, M. Merchant, R. Schwall, E.R. Neptune, Hepatocyte growth factor, a determinant of airspace homeostasis in the murine lung, PLoS Genet. 9 (2013) e1003228. [7] A. Furlan, Z. Kherrouche, R. Montagne, M.C. Copin, D. Tulasne, Thirty years of research on Met receptor to move a biomarker from bench to bedside, Cancer Res. 74 (2014) 6737–6744. [8] D.R. Spigel, T.J. Ervin, R.A. Ramlau, D.B. Daniel, J.H. Goldschmidt Jr., G.R. Blumenschein Jr., M.J. Krzakowski, G. Robinet, B. Godbert, F. Barlesi, R. Govindan, T. Patel, S.V. Orlov, M.S. Wertheim, W. Yu, J. Zha, R.L. Yauch, P.H. Patel, S.C. Phan, A.C. Peterson, Randomized phase II trial of onartuzumab in combination with erlotinib in patients with advanced non-small-cell lung cancer, J. Clin. Oncol. 31 (2013) 4105–4114. [9] R. Dziadziuszko, M.W. Wynes, S. Singh, B.R. Asuncion, J. Ranger-Moore, K. Konopa, W. Rzyman, B. Szostakiewicz, J. Jassem, F.R. Hirsch, Correlation between MET gene copy number by silver in situ hybridization and protein expression by immunohistochemistry in non-small cell lung cancer, J. Thorac. Oncol. 7 (2012) 340–347. [10] L. Schmidt, F.M. Duh, F. Chen, T. Kishida, G. Glenn, P. Choyke, S.W. Scherer, Z. Zhuang, I. Lubensky, M. Dean, R. Allikmets, A. Chidambaram, U.R. Bergerheim, J.T. Feltis, C. Casadevall, A. Zamarron, M. Bernues, S. Richard, C.J. Lips, M.M. Walther, L.C. Tsui, L. Geil, M.L. Orcutt, T. Stackhouse, B. Zbar, et al., Germline and somatic mutations in the tyrosine kinase domain of the MET proto-oncogene in papillary renal carcinomas, Nat. Genet. 16 (1997) 68–73. [11] Network Tcgar, Comprehensive molecular profiling of lung adenocarcinoma, Nature 511 (2014) 543–550. [12] G.M. Frampton, S.M. Ali, M. Rosenzweig, J. Chmielecki, X. Lu, T.M. Bauer, M. Akimov, J.A. Bufill, C. Lee, D. Jentz, R. Hoover, S.H. Ou, R. Salgia, T. Brennan, Z.R. Chalmers, S. Jaeger, A. Huang, J.A. Elvin, R. Erlich, A. Fichtenholtz, K.A. Gowen, J. Greenbowe, A. Johnson, D. Khaira, C. McMahon, E.M. Sanford, S. Roels, J. White, J. Greshock, R. Schlegel, D. Lipson, R. Yelensky, D. Morosini, J.S. Ross, E. Collisson, M. Peters, P.J. Stephens, V.A. Miller, Activation of MET via diverse exon 14 splicing alterations occurs in multiple tumor types and confers clinical sensitivity to MET inhibitors, Cancer Discov. 5 (2015) 850–859. [13] J.A. Engelman, K. Zejnullahu, T. Mitsudomi, Y. Song, C. Hyland, J.O. Park, N. Lindeman, C.M. Gale, X. Zhao, J. Christensen, T. Kosaka, A.J. Holmes, A.M. Rogers, F. Cappuzzo, T. Mok, C. Lee, B.E. Johnson, L.C. Cantley, P.A. Janne, MET

Author's Personal Copy M.-C. Copin et al. / Lung Cancer 101 (2016) 59–67

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling, Science 316 (2007) 1039–1043. S.H. Ou, E.L. Kwak, C. Siwak-Tapp, J. Dy, K. Bergethon, J.W. Clark, D.R. Camidge, B.J. Solomon, R.G. Maki, Y.J. Bang, D.W. Kim, J. Christensen, W. Tan, K.D. Wilner, R. Salgia, A.J. Iafrate, Activity of crizotinib (PF02341066), a dual mesenchymal-epithelial transition (MET) and anaplastic lymphoma kinase (ALK) inhibitor, in a non-small cell lung cancer patient with de novo MET amplification, J. Thorac. Oncol. 6 (2011) 942–946. R. Schwab, I. Petak, M. Kollar, F. Pinter, E. Varkondi, A. Kohanka, H. Barti-Juhasz, J. Schonleber, D. Brauswetter, L. Kopper, L. Urban, Major partial response to crizotinib, a dual MET/ALK inhibitor, in a squamous cell lung (SCC) carcinoma patient with de novo c-MET amplification in the absence of ALK rearrangement, Lung Cancer 83 (2014) 109–111. D.R. Camidge, S.I. Ou, G. Shapiro, G.A. Otterson, L.C. Villaruz, M.A. Villalona-Calero, A.J. Iafrate, M. Varella-Garcia, S. Dacic, S. Cardarella, W. Zhao, L. Tye, P. Stephenson, K.D. Wilner, L.P. James, M.A. Socinski, Efficacy and safety of crizotinib in patients with advanced C-MET amplified non-small cell lung cancer (NSCLC), ASCO Annu. Meeting J. Clin. Oncol. 32 (2014) 5s (2014 ASCO Annual Meeting. J Clin Oncol 32:5s, 2014. Abstract 8001 2014). P.K. Paik, A. Drilon, P.D. Fan, H. Yu, N. Rekhtman, M.S. Ginsberg, L. Borsu, N. Schultz, M.F. Berger, C.M. Rudin, M. Ladanyi, Response to MET inhibitors in patients with stage IV lung adenocarcinomas harboring MET mutations causing exon 14 skipping, Cancer Discov. 5 (2015) 842–849. X. Liu, Y. Jia, M.B. Stoopler, Y. Shen, H. Cheng, J. Chen, M. Mansukhani, S. Koul, B. Halmos, A.C. Borczuk, Next-generation sequencing of pulmonary sarcomatoid carcinoma reveals high frequency of actionable MET gene mutations, J. Clin. Oncol. 34 (2015) 794–802. P. Longati, A. Bardelli, C. Ponzetto, L. Naldini, P.M. Comoglio, Tyrosines1234-1235 are critical for activation of the tyrosine kinase encoded by the MET proto-oncogene (HGF receptor), Oncogene 9 (1994) 49–57. C. Ponzetto, A. Bardelli, Z. Zhen, F. Maina, P. dalla Zonca, S. Giordano, A. Graaziani, G. Panayotou, P.M. Comoglio, A multifunctional docking site mediates signaling and transformation by the hepatocyte growth factor/scatter factor receptor family, Cell 77 (1994) 261–271. Y. Nakamura, T. Niki, A. Goto, T. Morikawa, K. Miyazawa, J. Nakajima, M. Fukayama, c-Met activation in lung adenocarcinoma tissues: an immunohistochemical analysis, Cancer Sci. 98 (2007) 1006–1013. I. Watermann, B. Schmitt, F. Stellmacher, J. Muller, R. Gaber, C. Kugler, N. Reinmuth, R.M. Huber, M. Thomas, P. Zabel, K.F. Rabe, D. Jonigk, A. Warth, E. Vollmer, M. Reck, T. Goldmann, Improved diagnostics targeting c-MET in non-small cell lung cancer: expression, amplification and activation? Diagn. Pathol. 10 (2015) 130. J. Lefebvre, F. Ancot, C. Leroy, G. Muharram, A. Lemiere, D. Tulasne, Met degradation: more than one stone to shoot a receptor down, FASEB J. 26 (2012) 1387–1399. J. Lefebvre, G. Muharram, C. Leroy, Z. Kherrouche, R. Montagne, G. Ichim, S. Tauszig-Delamasure, A. Chotteau-Lelievre, C. Brenner, P. Mehlen, D. Tulasne, Caspase-generated fragment of the Met receptor favors apoptosis via the intrinsic pathway independently of its tyrosine kinase activity, Cell. Death Dis. 4 (2013) e871. R. Montagne, M. Berbon, L. Doublet, N. Debreuck, A. Baranzelli, H. Drobecq, C. Leroy, N. Delhem, H. Porte, M.C. Copin, E. Dansin, A. Furlan, D. Tulasne, Necrosis- and apoptosis-related Met cleavages have divergent functional consequences, Cell. Death Dis. 6 (2015) e1769. F. Ancot, C. Leroy, G. Muharram, J. Lefebvre, J. Vicogne, A. Lemiere, Z. Kherrouche, B. Foveau, A. Pourtier, O. Melnyk, S. Giordano, A. Chotteau-Lelievre, D. Tulasne, Shedding-generated Met receptor fragments can be routed to either the proteasomal or the lysosomal degradation pathway, Traffic 13 (2012) 1261–1272.

67

[27] B. Foveau, F. Ancot, C. Leroy, A. Petrelli, K. Reiss, V. Vingtdeux, S. Giordano, V. Fafeur, D. Tulasne, Downregulation of the Met receptor tyrosine kinase by presenilin-dependent regulated intramembrane proteolysis, Mol. Biol. Cell 20 (2009) 2494–2506. [28] V. Ludovini, S. Gori, M. Colozza, L. Pistola, E. Rulli, I. Floriani, E. Pacifico, F.R. Tofanetti, A. Sidoni, C. Basurto, A. Rulli, L. Crino, Evaluation of serum HER2 extracellular domain in early breast cancer patients: correlation with clinicopathological parameters and survival, Ann. Oncol. 19 (2008) 883–890. [29] C. Tse, A.S. Gauchez, W. Jacot, P.J. Lamy, HER2 shedding and serum HER2 extracellular domain: biology and clinical utility in breast cancer, Cancer Treat. Rev. 38 (2012) 133–142. [30] R. Paumelle, D. Tulasne, Z. Kherrouche, S. Plaza, C. Leroy, S. Reveneau, B. Vandenbunder, V. Fafeur, D. Tulashe, Hepatocyte growth factor/scatter factor activates the ETS1 transcription factor by a RAS-RAF-MEK-ERK signaling pathway, Oncogene 21 (2002) 2309–2319. [31] C. Simonneau, B. Leclercq, A. Mougel, E. Adriaenssens, C. Paquet, L. Raibaut, N. Ollivier, H. Drobecq, J. Marcoux, S. Cianférani, D. Tulasne, H. De Jonge, O. Melnyk, J. Vicogne, Semi-synthesis of a HGF/SF kringle one (K1) domain scaffold generates a potent in vivo MET receptor agonist, Chem. Sci. 6 (2015) 2110–2121. [32] L. Fu, W. Guo, B. Liu, L. Sun, Z. Bi, L. Zhu, X. Wang, B. Liu, Q. Xie, K. Li, Shedding of c-Met ectodomain correlates with c-Met expression in non-small cell lung cancer, Biomarkers 18 (2013) 126–135. [33] H. Lv, B. Shan, Z. Tian, Y. Li, Y. Zhang, S. Wen, Soluble c-Met is a reliable and sensitive marker to detect c-Met expression level in lung cancer, BioMed. Res. Int. 2015 (2015) 626578. [34] J. Madoz-Gurpide, S. Zazo, C. Chamizo, V. Casado, C. Carames, E. Gavin, I. Cristobal, J. Garcia-Foncillas, F. Rojo, Activation of MET pathway predicts poor outcome to cetuximab in patients with recurrent or metastatic head and neck cancer, J. Transl. Med. 13 (2015) 282. [35] J.G. Wu, J.W. Yu, H.B. Wu, L.H. Zheng, X.C. Ni, X.Q. Li, G.Y. Du, B.J. Jiang, Expressions and clinical significances of c-MET, p-MET and E2f-1 in human gastric carcinoma, BMC Res. Notes 7 (2014) 6. [36] K.P. Raghav, W. Wang, S. Liu, M. Chavez-MacGregor, X. Meng, G.N. Hortobagyi, G.B. Mills, F. Meric-Bernstam, G.R. Blumenschein Jr., A.M. Gonzalez-Angulo, cMET and phospho-cMET protein levels in breast cancers and survival outcomes, Clin. Cancer Res. 18 (2012) 2269–2277. [37] E. Arriola, I. Canadas, M. Arumi-Uria, M. Domine, J.A. Lopez-Vilarino, O. Arpi, M. Salido, S. Menendez, E. Grande, F.R. Hirsch, S. Serrano, B. Bellosillo, F. Rojo, A. Rovira, J. Albanell, MET phosphorylation predicts poor outcome in small cell lung carcinoma and its inhibition blocks HGF-induced effects in MET mutant cell lines, Br. J. Cancer 105 (2011) 814–823. [38] T. Onitsuka, H. Uramoto, K. Ono, M. Takenoyama, T. Hanagiri, T. Oyama, H. Izumi, K. Kohno, K. Yasumoto, Comprehensive molecular analyses of lung adenocarcinoma with regard to the epidermal growth factor receptor, K-ras, MET, and hepatocyte growth factor status, J. Thorac. Oncol. 5 (2010) 591–596. [39] K. Tsuta, Y. Kozu, T. Mimae, A. Yoshida, T. Kohno, I. Sekine, T. Tamura, H. Asamura, K. Furuta, H. Tsuda, c-MET/phospho-MET protein expression and MET gene copy number in non-small cell lung carcinomas, J. Thorac. Oncol. 7 (2012) 331–339. [40] G. Athauda, A. Giubellino, J.A. Coleman, C. Horak, P.S. Steeg, M.J. Lee, J. Trepel, J. Wimberly, J. Sun, A. Coxon, T.L. Burgess, D.P. Bottaro, c-Met ectodomain shedding rate correlates with malignant potential, Clin. Cancer. Res. 12 (2006) 4154–4162.