Supplementary Methods
Cell lines
Cell lines were obtained from American Type Culture Collection (Manassas, Virginia), National Cancer Institute Division of Cancer Treatment and Diagnosis Tumor Repository, Bethesda, Maryland; or Japan Health Sciences Foundation, Tokyo, Japan. Upon receipt, cells were immediately banked within the Genentech cell bank and utilized within limited passage. For authentication, STR profiles were determined for each line using the Promega PowerPlex 16 System. This was performed once and compared to external STR profiles of cell lines (when available) to determine cell line ancestry. Sixteen loci (fifteen STR loci and Amelogenin for gender identification) were analyzed including, D3S1358, TH01, D21S11, D18S51, Penta E, D5S818, D13S317, D7S820, D16S539, CSF1PO, Penta D, AMEL, vWA, D8S1179, and TPOX. Cell lines were grown in RPMI 1640 supplemented with 10% fetal bovine. To produce formalin-fixed paraffin-embedded sections, cell pellets were grown in eight T175 flasks for each cell line, dislodging cells with 10 mmol/L ethylenediaminetetraacetic acid (EDTA) (pH 8.0) followed by centrifugation, washing in phosphate-buffered saline, and then fixation overnight in 10% neutral buffered formalin. Fixed pellets were processed in a Tissue-Tek processor (Sakura Finetek, California) and then embedded in a paraffin block before construction of the tissue microarray (TMA).
H1155 and LXFL529 cell lines were engineered to stably express MET through lentiviral-mediated gene delivery. Lentivirus was produced by co-transfection of 293T cells with a lentiviral construct encoding the open reading frame of MET (GeneCopoeia, Rockville, Maryland) and lentiviral packaging plasmids, as per the Lenti-Pac HIV Expression Packaging kit (GeneCopoeia) instructions. Lentiviral particles were collected 48 hours after transfection and filtered through a 0.45 μM polyethersulfone filter before the addition of Polybrene (10 μg/mL). Target cells (5×104, plated 1 day before infection) were cultured with medium containing lentiviral particles for 48 hours before addition of puromycin (1 μg/mL) for selection of stably transduced cells.
H441 and EBC1 cell lines were engineered to express an inducible MET-targeting short-hairpin (sh)RNA through retroviral-mediated gene delivery. An oligonucleotide (5′―GATCCCCGAACAGAATCACTGACATATTCAAGAGATATGTCAGTGATTCTGTTCTTTTTTGGAAA―3′; bold text signifies the target hybridizing sequence) encoding an shRNA sequence against MET was cloned into BglII/HindIII sites of the pShuttle-H1 vector downstream of the H1 promoter. This plasmid was recombined with the retroviral pHUSH-GW vector using Clonase II enzyme (Invitrogen, Carlsbad, California), generating constructs in which shRNA expression was controlled by an inducible promoter (ref. S1). GP-293 packaging cells were co-transfected using FuGene 6 Transfection Reagent (Promega Corporation, Madison, Wisconsin) and CalPhos Mammalian Transfection Kits with the pVSV-G retroviral vector (both Clontech Laboratories, Mountain View, California) and the above recombinant retroviral constructs. Medium containing the recombinant virus was then added to EBC1 and H441 cells and recombinants were selected in puromycin.
A2780 cells stably overexpressing MST1R and matched vector control lines were a kind gift from Dr Amitabha Chaudhuri (Genentech, Inc., South San Francisco, California).
Gene expression in cell lines
Basal MET gene expression in NSCLC cell lines was assessed using RNA obtained from sub-confluent cell cultures on the Affymetrix (Santa Clara, California) microarray platform (HGU133Plus_2.0 chips). Preparation of complementary RNA and array hybridizations was carried out according to the manufacturer’s protocol. Expression data for MET (probe 207510_at) were determined using the robust multichip average normalization algorithm. The significance of the relationship between MET IHC scores and mRNA expression was determined using the Jonckheere–Terpstra test.
Flow cytometry
Determination of MET cell surface expression levels was carried out by flow cytometry. Cells (1×106) were directly stained with the 5D5 monoclonal antibody or a murine isotype control antibody, washed, resuspended in 1 mL buffer containing 10 μg/mL propidium iodide, and live cells were subsequently analyzed for mean shift in fluorescence on a FACSCalibur (Becton Dickinson, Franklin Lakes, New Jersey). Normalized mean fluorescence intensity (MFI) was calculated (5D5MFI – control IgMFI).
MET expression by IHC
MET expression levels were evaluated in archival tissue specimens using CONFIRM anti-total MET (SP44) rabbit monoclonal primary antibody (Ventana Medical Systems, Inc.; cat no. 790-4430), according to the manufacturer’s instructions. Staining was performed on the Ventana Benchmark XT instrument using CC1 standard antigen retrieval (Ventana Medical Systems). Incubation with primary antibody was performed for 16 minutes at 37°C using a 9.75µg/mL concentration of primary antibody. Specifically bound primary antibody was detected using ultraView methodology with diaminobenzidine (Ventana Medical Systems); sections were counterstained with hematoxylin. A composite scoring system (0, 1, 2, or 3) was devised that evaluated both staining intensity (negative, weak, moderate, and strong), and the proportion of tumor cells that exhibited the respective staining intensity). Tumors with ≥50% of tumor cells exhibiting moderate to strong staining intensity were predefined as MET-positive, prior to unblinding the treatment assignment. The scoring criteria are detailed in the Supplementary Table S3. The scoring of OAM4558g specimens was independently reviewed by a second pathologist, with an 88.3% concordance in calling MET-positive between pathologists. We did not observe a difference in outcome based upon pathologist #2’s interpretation of the data (OS HR=0.36, p=0.002 for pathologist #1 and OS HR=0.42, p=0.006 for pathologist #2 in the Met-positive group upon retrospective evaluation of n=111 specimens).
MET and EGFR gene copy number by FISH
MET and EGFR gene copy numbers were evaluated by FISH using a 319 kb probe constructed from 3 bacterial artificial chromosome (BAC) clones that spanned the entire MET gene on 7q31.1, and a 327 kb probe constructed from 2 BAC clones spanning the EGFR gene on 7p11.2. The CEP7 centromere probe (Abbott Molecular) was used as a control in a 3-color assay run on a single slide. The entire slide was scored and average copy number/cell was calculated from a minimum of 60 to 70 nuclei. High-level MET amplification was defined as tight gene clusters of ≥15 copies of MET in ≥10% of tumor cells, or a MET:CEP7 ratio of ≥2. Multiple MET copy number cut-offs were evaluated; however, ≥5 copies of MET/cell were predefined as the criterion for FISH+, based on prior data supporting this cut-off as an independent negative prognostic factor in NSCLC (ref. S2). For analyses evaluating high MET and EGFR copy numbers, tumors were considered MET and EGFR FISH+ based on a scoring system developed by the University of Colorado, which has been utilized in multiple clinical studies evaluating EGFR small molecule inhibitors (ref. S3).
qRT-PCR profiling of EGFR and MET, and related mRNA transcripts from clinical specimens
RNA was isolated from tumor tissue slides that were macro-dissected to enrich for tumor content, if tumor content was below 75%. Expression of MET, HGF, EGFR, AREG, and EREG mRNA was evaluated by qRT-PCR on the Biomark platform (Fluidigm). The primer/probes utilized for profiling are shown in Supplementary Table S2. All primers/probes were previously tested for optimal performance on RNA extracted from formalin-fixed, paraffin-embedded tissue.
Additional tissue specimens for prevalence studies
Additional tissue specimens were collected to allow analysis of the prevalence of MET expression in NSCLC using IHC. The first set of tissue specimens, obtained from commercial vendors, was used to construct a TMA for NSCLC (TMA 1), without any patient-specific identifiers and clinical outcome information. The second specimen set (TMA 2) included primary NSCLCs (n = 446) that were surgically resected with curative intent at the MD Anderson Cancer Center (Houston, Texas) between Jan 2003 and Dec 2005.
Supplementary References
S1. Hoeflich KP, Gray DC, Eby MT, Tien JY, Wong L, Bower J, et al. Oncogenic BRAF is required for tumor growth and maintenance in melanoma models. Cancer Res 2006;66:999−1006.
S2. Cappuzzo F, Marchetti A, Skokan M, Rossi E, Gajapathy S, Felicioni L, et al. Increased MET gene copy number negatively affects survival of surgically resected non-small-cell lung cancer patients. J Clin Oncol 2009;27:1667−74.
S3. Varella-Garcia M, Diebold J, Eberhard DA, Geenen K, Hirschmann A, Kockx M, et al. EGFR fluorescence in situ hybridisation assay: guidelines for application to non-small-cell lung cancer. J Clin Pathol 2009;62:970−7.
Supplementary Figure Legends
Supplementary Figure S1. Characterization of the specificity of monoclonal antibody SP44. (A) Immunoreactivity of SP44 upon doxycycline-inducible knock-down of the MET gene in two NSCLC cell lines. (B) Immunoreactivity of SP44 in two NSCLC cell lines engineered to ectopically express MET. (C) Immunoreactivity of SP44 in a cell line engineered to ectopically express MST1R. (D) Immunohistochemistry staining with SP44 or an isotype control antibody of the two NSCLC cell lines engineered to ectopically express MET shown in Fig. 1B.
Supplementary Figure S2. (A) Immunohistochemistry staining of formalin-fixed paraffin-embedded NSCLC cell lines depicts a range of staining intensities. (B) Relationship of SP44 immunohistochemistry staining intensity with cell surface MET protein levels detected using monoclonal antibody 5D5 (bivalent form of onartuzumab) by flow cytometry. (C) Normalized MFI values (5D5MFI – control immunoglobulinMFI) across a panel of NSCLC, as determined in B above. Note that no binding of 5D5 was observed in cell lines staining negative for SP44.
Supplementary Figure S3. Relationship of SP44 IHC staining intensity with MET mRNA levels across a panel of NSCLC cell lines. mRNA levels were determined on gene expression arrays and values represent robust multi-array average normalized data.
Supplementary Figure S4. (A) Examples of SP44 IHC staining patterns in human lung tissue. (B) Representative images of SP44 staining of benign lung tissues depicting reactivity in respiratory mucosa and pneumocytes.
Supplementary Figure S5. Relationship of H-score with the clinical scoring metric.
Supplementary Figure S6. Detailed breakdown of SP44 staining intensity in NSCLC specimens with H-score.
Supplementary Figure S7. Relationship of MET copy number with MET IHC clinical score.
Supplementary Figure S8. Kaplan–Meier estimates of OS according to MET copy number in the ITT and EGFRwt population. (A) ITT patients with tumors with ≥4 copies of MET/cell as determined by FISH (n=30). (B) ITT patients with tumors with ≥3 copies of MET/cell as determined by FISH (n=65). (C) EGFRwt patients with tumors with ≥4 copies of MET/cell as determined by FISH (n=25). (D) EGFRwt patients with tumors with ≥3 copies of MET/cell as determined by FISH (n=56). (E) Kaplan–Meier estimates of OS in MET and EGFR FISH+ patients in the ITT population.
Supplementary Figure S9. Baseline plasma HGF levels based on treatment arm and diagnostic subgroup.
8