Histopathology 2014, 65, 879–896. DOI: 10.1111/his.12510
Immunohistochemical application of a highly sensitive and specific murine monoclonal antibody recognising the extracellular domain of the human hepatocyte growth factor receptor (MET) Aaron M Gruver, Ling Liu, Peter Vaillancourt, Sau-Chi B Yan, Joel D Cook, Jessica A Roseberry Baker, Erin M Felke, Megan E Lacy, Christophe C Marchal, Hadrian Szpurka, Timothy R. Holzer, Emily K Rhoads, Wei Zeng, Mark A Wortinger, Jirong Lu, Chi-kin Chow, Irene J Denning, Gregory Beuerlein, Julian Davies, Jeff C Hanson, Kelly M Credille, Sameera R Wijayawardana & Andrew E Schade Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN, USA Date of submission 24 April 2014 Accepted for publication 12 July 2014 Published online Article Accepted 16 July 2014
Gruver A M, Liu L, Vaillancourt P, Yan S-C B, Cook J D, Roseberry Baker J A, Felke E M, Lacy M E, Marchal C C, Szpurka H, Holzer T R, Rhoads E K, Zeng W, Wortinger M A, Lu J, Chow C-K, Denning I J, Beuerlein G, Davies J, Hanson J C, Credille K M, Wijayawardana S R, Schade A E (2014) Histopathology 65, 879–896
Immunohistochemical application of a highly sensitive and specific murine monoclonal antibody recognising the extracellular domain of the human hepatocyte growth factor receptor (MET) Aims: Development of novel targeted therapies directed against hepatocyte growth factor (HGF) or its receptor (MET) necessitates the availability of quality diagnostics to facilitate their safe and effective use. Limitations of some commercially available anti-MET antibodies have prompted development of the highly sensitive and specific clone A2H2-3. Here we report its analytical properties when applied by an automated immunohistochemistry method. Methods and results: Excellent antibody specificity was demonstrated by immunoblot, ELISA, and IHC evaluation of characterised cell lines including NIH3T3 overexpressing the related kinase MST1R
(RON). Sensitivity was confirmed by measurements of MET in cell lines or characterised tissues. IHC correlated well with FISH and quantitative RT-PCR assessments of MET (P < 0.001). Good total agreement (89%) was observed with the anti-MET antibody clone SP44 using whole-tissue sections, but poor positive agreement (21–47%) was seen in tissue microarray cores. Multiple lots displayed appropriate reproducibility (R2 > 0.9). Prevalence of MET positivity by IHC was higher in non-squamous cell NSCLC, MET or EGFR amplified cases, and in tumours harbouring abnormalities in EGFR exon 19 or 21.
Abbreviations: DAB, diaminobenzidine; ECD, extracellular domain; EGFR, epidermal growth factor receptor; ELISA, enzyme linked immunosorbent assay; ELISA, enzyme-linked immunosorbent assay; Fab, fragment antigen binding; FFPE, formalin fixed paraffin-embedded; FISH, fluorescence in situ hybridization; H&E, hematoxylin and eosin; HGF, hepatocyte growth factor; HRP, horseradish peroxidase; IHC, immunohistochemistry; KS-test, Kolmogorov-Smirnov test; mAb, monoclonal antibody; MACS, magnetic activated cell-sorting; MET-IR, MET immunoreactivity; NSCLC, non-small cell lung cancer; OD, optical density; RT, room temperature; RT-PCR, real time polymerase chain reaction; SCC, squamous cell carcinoma; TBST, tris buffered saline w/Tween-20; TMA, tissue microarray; TMB, tetramethylbenzidine. Address for correspondence: A M Gruver, MD, PhD, Medical Advisor, Diagnostic and Experimental Pathology, Eli Lilly and Company, Lilly Corporate Center, 307 E. McCarty Street, Indianapolis, IN 46285, USA. e-mail: [email protected] © 2014 The Authors. Histopathology Published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
880
A M Gruver et al.
Conclusions: The anti-MET antibody clone A2H2-3 displays excellent specificity and sensitivity. These properties make it suitable for clinical trial investiga-
tions and development as a potential companion diagnostic.
Keywords: c-MET, hepatocyte growth factor, immunohistochemistry, LY2875358, MET, predictive biomarker
Materials and methods
Introduction The proto-oncogene MET encodes a receptor tyrosine kinase that, after binding with its ligand, hepatocyte growth factor (HGF), activates a wide range of different cellular signalling pathways including those involved in proliferation, motility, migration, angiogenesis and invasion.1 Although MET is important in the control of tissue homeostasis under normal physiological conditions including development, it has also been found to be aberrantly activated in human cancers via mutation, amplification, protein overexpression or overexpression of its ligand.2 Several novel therapies directed against MET or HGF are under clinical investigation in multiple tumour types.3–6 Development of analytically validated diagnostics to support the testing of tailoring hypotheses is foundational to clinical trial investigations for targeted therapies. Once an assay has been validated analytically, additional challenges come in determining the appropriate use and interpretation of the test prior to implementation in a clinical trial setting. Before investigating clinical utility, established antibody validation algorithms recommend comparing the immunohistochemistry (IHC) result to biologically proven controls usually assessed by alternate methodologies as part of the analytic validation.7 In-situ hybridization techniques for assessment of MET gene copy number and measurements of MET mRNA expression are used as comparators, and the correlation between these alternate methods and the immunoreactivity produced by IHC has been investigated using commercially available antibodies.8–11 Existing MET antibodies have demonstrated significant lot-tolot variability, and the limitations associated with some available clones have generated interest in the development of novel multipurpose monoclonal antibodies (mAbs) recognising MET.12,13 Here we describe the testing of a unique mAb that recognises the extracellular domain of human MET with high sensitivity and specificity. Experiments were conducted to determine whether the A2H2-3 antibody clone, when applied using commercially available IHC reagents and equipment, demonstrates satisfactory analytical performance.
TISSUE AND CELL LINES
Formalin-fixed, paraffin-embedded tissue (FFPE) blocks were acquired from commercial sources. In total 271 non-small cell lung carcinoma (NSCLC) tumours comprised of 109 specimens from TMA LC121, 95 specimens from TMA LC1922 (both acquired from US Biomax, Rockville, MD, USA), and 67 NSCLC whole-tissue specimens purchased from Asterand (Detroit, MI, USA), US Biomax, Capital Biosciences, Cybridi (Gaithersburg, MD, USA), Indivumed (Baltimore, MD, USA), Conversant Bio (Huntsville, AL, USA) and Indiana University (Bloomington, IN, USA) were analysed. Twenty late-stage gastric carcinoma specimens were purchased from Asterand and US Biomax. Quality control was performed on obtained tissues to confirm diagnosis and antigen preservation. Available tissue blocks were used for tissue microarray (TMA) construction and preparation of whole sections, unless only pre-cut unstained material was available. TMA LYSTM-2 contains 20 specimens of gastric carcinoma sampled in duplicate 1 mm cores (10 tubular adenocarcinoma, seven poorly cohesive carcinoma, one mucinous carcinoma, and two mixed carcinoma). TMA LYLNG-2 contains 47 specimens of late stage, NSCLC (21 adenocarcinoma, 15 squamous cell carcinoma, five large cell carcinoma, four adenosquamous carcinoma, and two poorly differentiated carcinoma, not otherwise specified) sampled in duplicate 1 mm cores. Three of the tumours on LC121 and 1 of the tumours on LC1922 were reassigned classification following assessment of TTF-1 and p63 immunostaining (Clarient Inc., Aliso Viejo, CA, USA) using an algorithm validated for small specimens,14 and one core on both TMAs lacked tumour; therefore, for this study specimens analysed from LC121 consisted of 22 squamous cell carcinoma, 37 large cell carcinoma, and 50 adenocarcinoma. The specimens from LC1922 consisted of 49 squamous cell carcinoma, nine adenosquamous carcinoma, 36 adenocarcinoma, and one large cell carcinoma. The A549, DMS114, H1299, H1650, H1975, H1993, H2444, H441, H520, HCC827, HS746T,
© 2014 The Authors. Histopathology Published by John Wiley & Sons Ltd, Histopathology, 65, 879–896.
890
A M Gruver et al.
A
B
C
D
E
F
G
H
I
J
K
L
Figure 6. Representative images captured from NSCLC tested by the IHC assay using multiple lots of the A2H2-3 antibody. Lot 1 (A, D, G, J), Lot 2 (B, E, H, K), and Lot 3 (C, F, I, L).
PREVALENCE OF MET PROTEIN EXPRESSION BY DISEASE CLASSIFICATION
Once satisfactory performance of the IHC method in FFPE tissue was demonstrated, an additional 204 NSCLC tumours were tested to understand the prevalence of MET expression by IHC within the context of the histologic and molecular classifications of the
disease. Summary findings from all specimens tested are provided in Table 5. MET positive cases were observed more often in: non-squamous cell NSCLC compared to squamous cell carcinoma (49% versus 17%), MET amplified cases compared to unamplified (100% versus 36%), EGFR amplified cases compared to unamplified (71% versus 41%) and in tumours harbouring abnormalities in EGFR exon 19 or 21
© 2014 The Authors. Histopathology Published by John Wiley & Sons Ltd, Histopathology, 65, 879–896.
882
A M Gruver et al.
4331182). TaqManâ Fast Universal PCR Master Mix and fast qPCR methods were used according to manufacturer’s instructions. Samples were run in duplicates or triplicates using ViiA7TM instrument. Relative expression was calculated as follows: Expression (AU) = (40-METCq)/(40-ACTBCq).
DSP DNA FFPE Tissue Kit (Qiagen, # 60404) following manufacturer instructions. Extracted DNA was further used for ‘DNA sample assessment;’ samples that successfully passed quality control were used for mutation analysis following manufacturer’s instructions.
MUTATION ANALYSIS
FLUORESCENCE IN-SITU HYBRIDIZATION (FISH)
The therascreen EGFR RGQ PCR Kit (Qiagen, #870121) was used for qualitative detection of exon 19 deletions and exon 21 (L858R) substitution mutation of the epidermal growth factor receptor (EGFR) gene in DNA derived from the source FFPE NSCLC specimens sampled on the LYLNG-2 TMA that were available. Briefly, two 5 lm tumour sections were used for DNA extraction using QIAamp
To identify MET and EGFR amplification, FISH was performed using BAC clone RP11-163C9 (orange) for MET and EGFR (green) both from Empire Genomics (Buffalo, NY, USA) with CEP 7 (aqua) from Abbott Molecular (Des Plaines, IL, USA) as a control probe. Specimens amplified for MET or EGFR were those that demonstrated a MET/CEP 7 or EGFR/CEP7 ratio of ≥2.0 or ≥4 copies of the gene in ≥40% of cells.18
Non-reducing
Reducing
Non-reducing
170 kDa 145 kDa
SN
SK
MET
Reducing
U H -1 52 0 SN U M 16 KN 45 SN U H -1 52 SN 0 M U-1 KN 6 45
B H R3 12 H 99 44 1 SK B H R3 12 H 99 44 1
A
MET
170 kDa 145 kDa
GAPDH
B
Optical Densitometry 450 nm
GAPDH 3.0 2.5 2.0 A2H2-3 binding to plate-bound MET-Fc
1.5
A2H2-3 binding to plate-bound RON A2H2-3 binding to plate-bound PlexinA2
1.0
A2H2-3 binding to plate-bound RET
0.5 0.0 0.001 0.01 0.1 1 10 100 1000 10000 A2H2-3 (ng/mL)
Expression Au (MET/ACTB)
C
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
P < 0.001
Average H-Score < 100
Average H-Score ≥ 100
Figure 1. Sensitivity of the A2H2-3 antibody was analysed by Western blot A, with seven tumour cell lines under non-reducing (no bmercaptoethanol) or reducing (with b-mercaptoethanol) conditions. Specificity of the A2H2-3 antibody was analysed by ELISA B, for METFc ECD (black circles) RON ECD (red triangles), plexin A2 (green squares) and RET ECD (blue diamonds). Correlation of the average H-score produced from the A2H2-3 IHC assay derived by image analysis to MET expression detected by quantitative RT-PCR in immortalised cell lines C. Horizontal black bar represents the median measurement of each group. The reported P-value corresponds to a twosample Kolmogorov-Smirnov test of equality of the cumulative distributions of the two groups.
© 2014 The Authors. Histopathology Published by John Wiley & Sons Ltd, Histopathology, 65, 879–896.
A2H2-3 anti-human MET antibody
STATISTICAL ANALYSIS
The two-sample Kolmogorov-Smirnov test (KS-test) was used to test whether the relative MET mRNA
883
expression levels corresponding to samples with an image-analysis derived average H-score
Immunohistochemical application of a highly sensitive and specific murine monoclonal antibody recognising the extracellular domain of the human hepatocyte growth factor receptor (MET) Aaron M Gruver, Ling Liu, Peter Vaillancourt, Sau-Chi B Yan, Joel D Cook, Jessica A Roseberry Baker, Erin M Felke, Megan E Lacy, Christophe C Marchal, Hadrian Szpurka, Timothy R. Holzer, Emily K Rhoads, Wei Zeng, Mark A Wortinger, Jirong Lu, Chi-kin Chow, Irene J Denning, Gregory Beuerlein, Julian Davies, Jeff C Hanson, Kelly M Credille, Sameera R Wijayawardana & Andrew E Schade Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN, USA Date of submission 24 April 2014 Accepted for publication 12 July 2014 Published online Article Accepted 16 July 2014
Gruver A M, Liu L, Vaillancourt P, Yan S-C B, Cook J D, Roseberry Baker J A, Felke E M, Lacy M E, Marchal C C, Szpurka H, Holzer T R, Rhoads E K, Zeng W, Wortinger M A, Lu J, Chow C-K, Denning I J, Beuerlein G, Davies J, Hanson J C, Credille K M, Wijayawardana S R, Schade A E (2014) Histopathology 65, 879–896
Immunohistochemical application of a highly sensitive and specific murine monoclonal antibody recognising the extracellular domain of the human hepatocyte growth factor receptor (MET) Aims: Development of novel targeted therapies directed against hepatocyte growth factor (HGF) or its receptor (MET) necessitates the availability of quality diagnostics to facilitate their safe and effective use. Limitations of some commercially available anti-MET antibodies have prompted development of the highly sensitive and specific clone A2H2-3. Here we report its analytical properties when applied by an automated immunohistochemistry method. Methods and results: Excellent antibody specificity was demonstrated by immunoblot, ELISA, and IHC evaluation of characterised cell lines including NIH3T3 overexpressing the related kinase MST1R
(RON). Sensitivity was confirmed by measurements of MET in cell lines or characterised tissues. IHC correlated well with FISH and quantitative RT-PCR assessments of MET (P < 0.001). Good total agreement (89%) was observed with the anti-MET antibody clone SP44 using whole-tissue sections, but poor positive agreement (21–47%) was seen in tissue microarray cores. Multiple lots displayed appropriate reproducibility (R2 > 0.9). Prevalence of MET positivity by IHC was higher in non-squamous cell NSCLC, MET or EGFR amplified cases, and in tumours harbouring abnormalities in EGFR exon 19 or 21.
Abbreviations: DAB, diaminobenzidine; ECD, extracellular domain; EGFR, epidermal growth factor receptor; ELISA, enzyme linked immunosorbent assay; ELISA, enzyme-linked immunosorbent assay; Fab, fragment antigen binding; FFPE, formalin fixed paraffin-embedded; FISH, fluorescence in situ hybridization; H&E, hematoxylin and eosin; HGF, hepatocyte growth factor; HRP, horseradish peroxidase; IHC, immunohistochemistry; KS-test, Kolmogorov-Smirnov test; mAb, monoclonal antibody; MACS, magnetic activated cell-sorting; MET-IR, MET immunoreactivity; NSCLC, non-small cell lung cancer; OD, optical density; RT, room temperature; RT-PCR, real time polymerase chain reaction; SCC, squamous cell carcinoma; TBST, tris buffered saline w/Tween-20; TMA, tissue microarray; TMB, tetramethylbenzidine. Address for correspondence: A M Gruver, MD, PhD, Medical Advisor, Diagnostic and Experimental Pathology, Eli Lilly and Company, Lilly Corporate Center, 307 E. McCarty Street, Indianapolis, IN 46285, USA. e-mail: [email protected] © 2014 The Authors. Histopathology Published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
880
A M Gruver et al.
Conclusions: The anti-MET antibody clone A2H2-3 displays excellent specificity and sensitivity. These properties make it suitable for clinical trial investiga-
tions and development as a potential companion diagnostic.
Keywords: c-MET, hepatocyte growth factor, immunohistochemistry, LY2875358, MET, predictive biomarker
Materials and methods
Introduction The proto-oncogene MET encodes a receptor tyrosine kinase that, after binding with its ligand, hepatocyte growth factor (HGF), activates a wide range of different cellular signalling pathways including those involved in proliferation, motility, migration, angiogenesis and invasion.1 Although MET is important in the control of tissue homeostasis under normal physiological conditions including development, it has also been found to be aberrantly activated in human cancers via mutation, amplification, protein overexpression or overexpression of its ligand.2 Several novel therapies directed against MET or HGF are under clinical investigation in multiple tumour types.3–6 Development of analytically validated diagnostics to support the testing of tailoring hypotheses is foundational to clinical trial investigations for targeted therapies. Once an assay has been validated analytically, additional challenges come in determining the appropriate use and interpretation of the test prior to implementation in a clinical trial setting. Before investigating clinical utility, established antibody validation algorithms recommend comparing the immunohistochemistry (IHC) result to biologically proven controls usually assessed by alternate methodologies as part of the analytic validation.7 In-situ hybridization techniques for assessment of MET gene copy number and measurements of MET mRNA expression are used as comparators, and the correlation between these alternate methods and the immunoreactivity produced by IHC has been investigated using commercially available antibodies.8–11 Existing MET antibodies have demonstrated significant lot-tolot variability, and the limitations associated with some available clones have generated interest in the development of novel multipurpose monoclonal antibodies (mAbs) recognising MET.12,13 Here we describe the testing of a unique mAb that recognises the extracellular domain of human MET with high sensitivity and specificity. Experiments were conducted to determine whether the A2H2-3 antibody clone, when applied using commercially available IHC reagents and equipment, demonstrates satisfactory analytical performance.
TISSUE AND CELL LINES
Formalin-fixed, paraffin-embedded tissue (FFPE) blocks were acquired from commercial sources. In total 271 non-small cell lung carcinoma (NSCLC) tumours comprised of 109 specimens from TMA LC121, 95 specimens from TMA LC1922 (both acquired from US Biomax, Rockville, MD, USA), and 67 NSCLC whole-tissue specimens purchased from Asterand (Detroit, MI, USA), US Biomax, Capital Biosciences, Cybridi (Gaithersburg, MD, USA), Indivumed (Baltimore, MD, USA), Conversant Bio (Huntsville, AL, USA) and Indiana University (Bloomington, IN, USA) were analysed. Twenty late-stage gastric carcinoma specimens were purchased from Asterand and US Biomax. Quality control was performed on obtained tissues to confirm diagnosis and antigen preservation. Available tissue blocks were used for tissue microarray (TMA) construction and preparation of whole sections, unless only pre-cut unstained material was available. TMA LYSTM-2 contains 20 specimens of gastric carcinoma sampled in duplicate 1 mm cores (10 tubular adenocarcinoma, seven poorly cohesive carcinoma, one mucinous carcinoma, and two mixed carcinoma). TMA LYLNG-2 contains 47 specimens of late stage, NSCLC (21 adenocarcinoma, 15 squamous cell carcinoma, five large cell carcinoma, four adenosquamous carcinoma, and two poorly differentiated carcinoma, not otherwise specified) sampled in duplicate 1 mm cores. Three of the tumours on LC121 and 1 of the tumours on LC1922 were reassigned classification following assessment of TTF-1 and p63 immunostaining (Clarient Inc., Aliso Viejo, CA, USA) using an algorithm validated for small specimens,14 and one core on both TMAs lacked tumour; therefore, for this study specimens analysed from LC121 consisted of 22 squamous cell carcinoma, 37 large cell carcinoma, and 50 adenocarcinoma. The specimens from LC1922 consisted of 49 squamous cell carcinoma, nine adenosquamous carcinoma, 36 adenocarcinoma, and one large cell carcinoma. The A549, DMS114, H1299, H1650, H1975, H1993, H2444, H441, H520, HCC827, HS746T,
© 2014 The Authors. Histopathology Published by John Wiley & Sons Ltd, Histopathology, 65, 879–896.
890
A M Gruver et al.
A
B
C
D
E
F
G
H
I
J
K
L
Figure 6. Representative images captured from NSCLC tested by the IHC assay using multiple lots of the A2H2-3 antibody. Lot 1 (A, D, G, J), Lot 2 (B, E, H, K), and Lot 3 (C, F, I, L).
PREVALENCE OF MET PROTEIN EXPRESSION BY DISEASE CLASSIFICATION
Once satisfactory performance of the IHC method in FFPE tissue was demonstrated, an additional 204 NSCLC tumours were tested to understand the prevalence of MET expression by IHC within the context of the histologic and molecular classifications of the
disease. Summary findings from all specimens tested are provided in Table 5. MET positive cases were observed more often in: non-squamous cell NSCLC compared to squamous cell carcinoma (49% versus 17%), MET amplified cases compared to unamplified (100% versus 36%), EGFR amplified cases compared to unamplified (71% versus 41%) and in tumours harbouring abnormalities in EGFR exon 19 or 21
© 2014 The Authors. Histopathology Published by John Wiley & Sons Ltd, Histopathology, 65, 879–896.
882
A M Gruver et al.
4331182). TaqManâ Fast Universal PCR Master Mix and fast qPCR methods were used according to manufacturer’s instructions. Samples were run in duplicates or triplicates using ViiA7TM instrument. Relative expression was calculated as follows: Expression (AU) = (40-METCq)/(40-ACTBCq).
DSP DNA FFPE Tissue Kit (Qiagen, # 60404) following manufacturer instructions. Extracted DNA was further used for ‘DNA sample assessment;’ samples that successfully passed quality control were used for mutation analysis following manufacturer’s instructions.
MUTATION ANALYSIS
FLUORESCENCE IN-SITU HYBRIDIZATION (FISH)
The therascreen EGFR RGQ PCR Kit (Qiagen, #870121) was used for qualitative detection of exon 19 deletions and exon 21 (L858R) substitution mutation of the epidermal growth factor receptor (EGFR) gene in DNA derived from the source FFPE NSCLC specimens sampled on the LYLNG-2 TMA that were available. Briefly, two 5 lm tumour sections were used for DNA extraction using QIAamp
To identify MET and EGFR amplification, FISH was performed using BAC clone RP11-163C9 (orange) for MET and EGFR (green) both from Empire Genomics (Buffalo, NY, USA) with CEP 7 (aqua) from Abbott Molecular (Des Plaines, IL, USA) as a control probe. Specimens amplified for MET or EGFR were those that demonstrated a MET/CEP 7 or EGFR/CEP7 ratio of ≥2.0 or ≥4 copies of the gene in ≥40% of cells.18
Non-reducing
Reducing
Non-reducing
170 kDa 145 kDa
SN
SK
MET
Reducing
U H -1 52 0 SN U M 16 KN 45 SN U H -1 52 SN 0 M U-1 KN 6 45
B H R3 12 H 99 44 1 SK B H R3 12 H 99 44 1
A
MET
170 kDa 145 kDa
GAPDH
B
Optical Densitometry 450 nm
GAPDH 3.0 2.5 2.0 A2H2-3 binding to plate-bound MET-Fc
1.5
A2H2-3 binding to plate-bound RON A2H2-3 binding to plate-bound PlexinA2
1.0
A2H2-3 binding to plate-bound RET
0.5 0.0 0.001 0.01 0.1 1 10 100 1000 10000 A2H2-3 (ng/mL)
Expression Au (MET/ACTB)
C
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
P < 0.001
Average H-Score < 100
Average H-Score ≥ 100
Figure 1. Sensitivity of the A2H2-3 antibody was analysed by Western blot A, with seven tumour cell lines under non-reducing (no bmercaptoethanol) or reducing (with b-mercaptoethanol) conditions. Specificity of the A2H2-3 antibody was analysed by ELISA B, for METFc ECD (black circles) RON ECD (red triangles), plexin A2 (green squares) and RET ECD (blue diamonds). Correlation of the average H-score produced from the A2H2-3 IHC assay derived by image analysis to MET expression detected by quantitative RT-PCR in immortalised cell lines C. Horizontal black bar represents the median measurement of each group. The reported P-value corresponds to a twosample Kolmogorov-Smirnov test of equality of the cumulative distributions of the two groups.
© 2014 The Authors. Histopathology Published by John Wiley & Sons Ltd, Histopathology, 65, 879–896.
A2H2-3 anti-human MET antibody
STATISTICAL ANALYSIS
The two-sample Kolmogorov-Smirnov test (KS-test) was used to test whether the relative MET mRNA
883
expression levels corresponding to samples with an image-analysis derived average H-score
