The Prognostic Significance of c-MET and EGFR Overexpression in Resected Gastric Adenocarcinomas.

ORIGINAL ARTICLE

The Prognostic Significance of c-MET and EGFR Overexpression in Resected Gastric Adenocarcinomas Aleksandra Paliga, MD,* Horia Marginean, MD,w Basile Tessier-Cloutier, MD,z Bibianna Purgina, MD,* Derek Jonker, MD,w and Esmeralda C. Marginean, MD*

Objectives: Epidermal growth factor receptor (EGFR) and c-MET are tyrosine kinase growth factor receptors implicated in gastric cancer (GC), and their pathways appear to be interdependent. The aim of this study was to investigate the prognostic value of EGFR and c-MET protein overexpression by immunohistochemistry in Canadian patients with resected GC and correlate it with clinicopathologic characteristics and overall survival (OS). Materials and Methods: Tissue microarray blocks were constructed from 120 resected GCs stained with EGFR and c-MET and scored semiquantitatively (0 to 3 +). Each receptor’s expression was compared with clinicopathologic characteristics and survival. Descriptive statistics, Kaplan–Meyer, and Cox regression were used for statistical analyses. Results: Of the 113 interpretable cases, overexpression of EGFR and c-MET was noted in 17 (15%) and 65 (57%), respectively; coexpression of EGFR and c-MET was observed in 12 (10%) of GC. EGFR and c-MET overexpression correlated with poor OS: median 13 versus 30 months in EGFR positive versus negative GC (hazard ratio [HR] = 1.67, P = 0.11); 27 versus 49 months in c-MET positive versus negative GC (HR = 1.17, P = 0.49), respectively. GC coexpressing EGFR and c-MET was significantly correlated with poor survival: 12 versus 29 months in double-positive versus rest of tumors both in univariate (HR = 2.62, P = 0.003) and multivariate analyses (HR = 2.58, P = 0.01). Conclusions: This study describes the prevalence and prognostic value of EGFR and c-MET in a Canadian population of patients undergoing curative intent resection for GC. Both c-MET and EGFR overexpression trended toward poor OS, but only the group with EGFR + / c-MET + GC reached statistical significance on multivariate analysis. Key Words: gastric cancer, EGFR, c-MET, immunohistochemistry, prognostic marker

(Am J Clin Oncol 2015;00:000–000)

From the *Department of Pathology and Laboratory Medicine, Division of Anatomical Pathology, The Ottawa Hospital, University of Ottawa; wThe Ottawa Hospital Research Institute, Ottawa, ON; and zDepartment of Pathology and Laboratory Medicine, The University of British Columbia, Vancouver, BC, Canada. Supported by the Ottawa Hospital Pathology and Laboratory Medicine (PALM) Academic Enrichment Fund. The authors declare no conflicts of interest. Reprints: Esmeralda C. Marginean, MD, Department of Pathology and Laboratory Medicine, Division of Anatomical Pathology, The Ottawa Hospital, University of Ottawa, CCW Rm 4251, 501 Smyth Rd, Ottawa, ON, Canada K1H 8L6. E-mail: [email protected]. Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www.amj clinicaloncology.com. Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0277-3732/15/000-000 DOI: 10.1097/COC.0000000000000202

American Journal of Clinical Oncology

G

astric cancer (GC) is the fourth most common cancer worldwide, affecting approximately 1 million people per year, and is the fifth most common malignancy in the world, after cancers of the lung, breast, colorectum, and prostate. This represents a substantive change since the very first estimates in 1975 when stomach cancer was the most common neoplasm.1–3 Half the world’s total cancer occurs in Eastern Asia (mainly in China). In Canada, incidence of stomach cancer is much lower than in other parts of the world like China, Eastern Asia, Eastern Europe, and some parts of South America and continues to decline in both males (2.3%/y) and females (1.3%/y), with estimated 3300 new cases per year (6.7 cases per 100,000). Lifetime probability of developing stomach cancer is estimated at 1.3% in males and 0.8% in females, which will be responsible for 3.2% of male deaths and 2.2% of female deaths in Canada (http://www.cancer.ca/˜/media/ cancer.ca/CW/cancer%20information/cancer%20101/Canadian %20cancer%20statistics/Canadian-Cancer-Statistics-2014--EN.pdf). Although our understanding of this disease has improved in the last decade, the prognosis for patients with advanced GC remains poor, with a median overall survival (OS) of 10% of tumor cells; 2 + = moderate membranous staining; 3 + = strong membranous staining, according to the published guidelines for HER2 in gastric/GEJ adenocarcinomas,46 and EGFR assessment in lung adenocarcinomas.47,48 The c-MET (cytoplasmic pattern) was graded similarly from 0 to 3 + , with 0 = no staining; 1 + = faint cytoplasmic staining in >10% of tumor cells; 2 + = moderate cytoplasmic staining; 3 + = strong cytoplasmic staining (Figs. 1A, B). An IHC score of 2 + and 3 + was considered positive/ overexpressed for both markers, with at least 10% of the tumor staining, whereas 1 + and 0 were considered negative. The highest score of the 4 tissue cores from each tumor was used as the final result. Discrepancies were resolved by review using a multiheaded microscope and consensus among pathologists.

Copyright

r

2015 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.


Volume 00, Number 00, ’’ 2015

Prognostic Significance of c-MET and EGFR

FIGURE 1. Immunohistochemical staining representative scores for EGFR and c-MET at 100 magnification.

Statistical Analysis Derivation of IHC Staining Intensity Thresholds (Positive Versus Negative) To determine the optimal cut points for dichotomization of each biomarker level (positive, negative), Cox regression coefficients comparison (hazard ratio [HR]) was applied to intensity of staining (0 to 3 +). The cut points were selected to maximize the prognostic ability of respective markers.

Survival Analysis The primary endpoint, OS, was defined as the time between the date of first pathologic diagnosis and the date of death due to any cause or was censored when last known alive. Survival was analyzed using the Kaplan-Meier method with log-rank testing to compare between groups. Cox proportional hazards regression modeling (Wald test) was used to assess the correlation between EGFR and c-MET intensity (positive, negative) and OS, with multivariate testing including the following as independent variables: sex, age (65 y and younger, older than 65 y), tumor location in the stomach (GEJ and cardia, body, antrum), histologic type (intestinal, diffuse, and mixed), tumor size, pT stage (1 to 4), pN stage (0 to 3), clinical stage (I, II, III, IV), surgical margins (R0, R1, R2), adjuvant chemotherapy, and adjuvant radiotherapy. The dichotomized biomarkers and variables with a P < 0.2 in univariate analysis were selected for inclusion in multivariable modeling. Proportional hazards assumptions were checked by including time-dependent covariates in the model. Median follow-up time was estimated based on the reverse Kaplan-Meier.49 The w2 test and Fisher exact test were used to test the statistical significance of differences in discrete data; Student t test and Mann-Whitney test were used to test the statistical significance of differences in continuous data. For all analyses, P < 0.05 was considered to be statistically significant unless otherwise stated. No adjustments were made for multiple comparisons. All analyses were performed with Stata 13 (StataCorp., College Station, TX). The report was written taking in consideration the Reporting of Tumor Marker Studies Guidelines.50

RESULTS Patient and Tumor Characteristics Of 120 available cases, 7 cases were excluded due to tissue loss at processing of TMA blocks, and 113 cases were Copyright

r

considered interpretable. The mean follow-up time was 80.0 months (range, 1 to 150 mo), with 6 patients lost to follow-up. The 5-year survival rate was 34% (95% confidence interval [CI], 25.3-43.6) and median OS was 29.0 months (95% CI, 21.7-47.9) (Fig. 2A). Pathologic and clinical characteristics for all GC patients are summarized in Table 1. Of the patients who were eligible for adjuvant chemotherapy, 49/113 (43%) received the routine standard of care adjuvant chemotherapy available at the time of resection, that is, 5-fluorouracil-based chemotherapy. Four of those 49 (8%) patients also received neoadjuvant 5-fluorouracil-based chemotherapy. No patients received any targeted therapies against EGFR or c-MET receptors.

IHC Results EGFR and c-MET protein expression was determined by IHC for the 113 GCs. For EGFR, 81 (72%) scored 0; 14 (12%) scored 1 + ; 4 (4%) scored 2 + ; and 13 (12%) scored 3 + . For cMET, 10 (8.8%) scored 0; 38 (34%) scored 1 + ; 38 (34%) scored 2 + ; and 27 (24%) scored 3 + . Overall, EGFR was positive in 17 (15%) cases and c-MET in 65 (58%) tumors. Coexpression of EGFR and c-MET was seen in 12 (10.6%) tumors (see Figure, Supplemental Digital Content 1, http:// links.lww.com/AJCO/A101, which demonstrates a Venn diagram of EGFR and c-MET coexpression). Neither EGFR nor c-MET positivity showed any association with pathologic T or N stage, location, histology, margin status, local recurrence, age, or sex.

Survival In univariate analyses, tested in a Cox proportional hazards model,51 the relationship between EGFR and c-MET expression, patient and tumor characteristics, and OS is summarized in Table 2. For statistical analysis, the pure diffuse type GC (n = 17) and mixed type GC, showing both diffuse and intestinal morphology (n = 10), were analyzed together because the mixed tumors showed a similar OS with the pure diffuse tumors. The following variables reached a significant level of 20% (P = 0.20) for correlation with OS and were subsequently used in the multivariate model: tumor location (P = 0.01), with gastric antral tumors having the best OS (79 mo) versus cardia (26 mo) and body (23 mo); pT stage (P = 0.00001); pN stage (P = 0.01); clinical stage (P = 0.00001); margin status (P = 0.001) (see Figure, Supplemental Digital Content 2, http://


www.amjclinicaloncology.com |

3


Paliga et al



FIGURE 2. A, Overall survival (OS)—primary analysis of all patients (B) OS by EGFR status (C) OS by c-MET status (D) OS in EGFR gastric cancer by c-MET status. E, OS in EGFR + gastric cancer by c-MET status (F) OS by double EGFR/c-MET positive versus rest. CI indicates confidence interval; EGFR, epidermal growth factor receptor; HR, hazard ratio; t, time.

links.lww.com/AJCO/A102, which demonstrates KaplanMeier survival curves for all these variables). Regarding the relationship between EGFR and c-MET status and OS in univariate analysis, there was a trend toward worse survival in EGFR + GC, with median survival of 15 versus 31 months in EGFR GC (HR = 1.60; 95% CI, 0.892.86; P = 0.11) (Fig. 2B). Similarly, c-MET GC trended

4 | www.amjclinicaloncology.com

toward worse survival, with median survival of 28 versus 48 months for positive versus negative c-MET status patients, respectively (HR = 1.17; 95% CI, 0.73-1.87; P = 0.49) (Fig. 2C). The interaction between EGFR and c-MET was explored. Among EGFR patients, c-MET status has no impact on the OS (Fig. 2D). Conversely, among EGFR + patients, overexpression of c-MET trended to a marked

Copyright

r





DISCUSSION

TABLE 1. Patient and Tumor Characteristics (N = 113)

Patient Characteristics Male Female Mean age (range) Clinical stage (%)* I A, B II III A, B IV Adjuvant chemotherapy Adjuvant radiotherapy Early metastasesw Late metastasesz Follow-up time (median [95% CI]) (mo) OS (median [95% CI]) (mo) Tumor characteristics pT stage T1 T2 a, b T3 T4 pN stage N0 N1 N2 N3 Tumor location Cardia Body Antrum/pylorus Histologic type Intestinal Diffuse/mixed Margins Positive Negative

Cases (n [%]) 81 (72) 32 (28) 64 (30-94) 23 (20) 31 (27) 44 (39) 15 (13) 50 (44) 47 (42) 17 (15) 50 (44) 80 (73-93) 29 (22-48) 11 38 51 13

(10) (34) (45) (12)

30 51 26 6

(27) (45) (23) (5)

32 (28) 47 (42) 34 (30) 86 (76) 27 (24) 39/2 (36) 72 (64)

*Tumors were staged at time of diagnosis according to Cancer Staging Manual, AJCC 2002 (6th edition) criteria. wDocumented metastases by imaging within 3 months postsurgery date (ie, before adjuvant therapy initiated). zMetastasis detected after 3 months postsurgery date. CI indicates confidence interval.

negative impact on OS, with median survival of 13 versus 79 months (HR = 3.45; 95% CI, 0.89-13.42, P = 0.07), for c-MET positive versus negative, respectively (Fig. 2E). In multivariate analyses, summarized in Table 3, EGFR status improved its status as a poor prognostic factor, however, without being statistically significant (P = 0.06), while c-MET status was still not significant (P = 0.46). Pathologic T and N stage maintained their significance, while tumor location and clinical stage lost their significant OS correlation. The prognostic value of EGFR/c-MET doublepositive signature, improved after adjustment for the other covariates, supporting the role of this biomarker combination in predicting OS of these patients. The risk of death for patients with EGFR + /c-MET + tumors was approximately triple that of the rest of the patients: HR = 2.62; 95% CI, 1.37-5.03; P = 0.002 (Fig. 2F). The overall prognostic accuracy (discriminatory power) of the final model was acceptable, as suggested by the high concordance index value (0.74).52,53 Possible departures from model assumptions were examined using Schoenfeld residuals.54 Copyright

r


The present study retrospectively evaluated the correlation of EGFR and c-MET expression by IHC with clinicopathologic characteristics and survival in resected gastric adenocarcinomas of 113 Canadian patients. The published frequency of EGFR positivity/overexpression by IHC in GC varies between 2% and 38%,15,55–57 whereas of c-MET varies between 9% and 83%.23,28–34,58–60 The varied reported range of EGFR and c-MET overexpression likely account from differing methodologies (TMA vs. whole section, biopsy vs. resection, primary tumor vs. metastasis), lack of standardized interpretation methods and possible geographic differences, with a predominance of studies investigating Asian populations. In our study, EGFR positivity was found in 15% of cases, and c-MET in 58% of cases, which is within the range of previous reports. EGFR positivity in our study trended to worse survival on univariate and multivariate analysis, concordant with the published literature.11,12,15,37,60–63 c-MET status alone had no statistical significance as a prognostic marker in our patient cohort, but did appear to have a negative impact on EGFR + tumors, supporting possible crosstalk between EGFR and c-MET. Despite varied methodologies and antibodies, most studies have reported c-MET positivity/ overexpression as a negative prognostic factor.29,31,32,64–68 Only 2 other studies have reported no c-MET prognostic significance,30,69 and 1 study has reported it as a positive prognostic factor.37 It is possible that with a larger patient series, the trend we saw toward negative survival would achieve statistical significance. The majority of published studies analyzing EGFR by IHC used a score of 2 + and 3 + to define positivity,13,15,33,38–41 a few used only a score of 3 + ,59,60 whereas others used a score combining intensity and percentage of staining.58 Since no consensus definition exists regarding what should be considered positive, we explored which level of dichotomization best predicted survival, and we reached the conclusion that cases with IHC score of 2 + and 3 + should be considered positive/ overexpressed. Similarly, no consensus exists on criteria to stratify c-MET expression. In 2 recent systematic reviews and meta-analyses, c-MET was evaluated by various methods, including IHC, quantitative reverse transcription polymerase chain reaction, florescence in situ hybridization, silver in situ hybridization (SISH), and Southern blot.34,70 The studies that evaluated c-MET by IHC were heterogeneous, either calculating composite scores from percentage and intensity, or looking at percentage alone, or staining intensity alone, with 2 + and 3 + being considered overexpressed.22,37 As with EGFR, we explored which level of dichotomization best predicted survival, and we reached the conclusion that tumors with c-MET IHC score of 2 + and 3 + should be considered positive/overexpressed (cytoplasmic and membranous staining). We feel that the varied published rates of EGFR and cMET overexpression relate to the use of different antibody clones, particularly when assessing c-MET. The main EGFR antibodies used in previous reports were clone 31G7 (Zymed Laboratories Inc., San Francisco, CA) or PharmDx kit (Dako Cytomation), which show comparable results.71 Studies on cMET are more heterogeneous, and evaluate its expression either by IHC, FISH, SISH, Southern blot, or quantitative reverse transcription polymerase chain reaction. Numerous antibodies against c-MET are available, including rabbit monoclonal C-28 (Santa Cruz Biotechnology, Santa Cruz, CA),72,73 phospho-MET (Y1234/Y1235; Cell Signaling Technology, Beverly, MA),32 MET (3D4) mouse mAb (Invitrogen, San



5



Paliga et al


TABLE 2. Association Between EGFR and c-MET, Patient and Tumor Characteristics, and OS in Gastric Cancer Patients; Univariate Analyses

Univariate Analysis

No. Patients (N = 113)

Median OS (95% CI) (mo)

96 17

30.5 (22.7-56.3) 15.0 (8.5-54.1)

48 65

47.9 (18.4-63.0) 28.3 (19.9-41.1)

101 12

34.9 (23.1-56.3) 12.6 (7.7-25.6)

32 81

37.8 (14.7-’) 25.9 (20.8-51.9)

21 20 33 25 14

28.3 45.3 30.3 29.0 15.2

EGFR status Negative (reference) Positive c-MET status Negative (reference) Positive EGFR/c-MET Rest (reference) EGFR + /c-MET + Sex Female (reference) Male Age 18-49 (reference) 50-59 60-69 70-79 80 + Age 65 + < 65 (reference) Z65 Location Antrum (reference) GEJ Body Histologic type Intestinal (reference) Diffuse/mixed T stage 1 (reference) 2 3 4 N stage 0 (reference) 1 2 3 Clinical stage I (reference) II II IV Resection margins R1 or R2 (reference) R0

HR (95% CI)

P 0.11

1.60 (0.89-2.87) 0.50 1.17 (0.74-1.87) 0.003 2.62 (1.37-5.03) 0.50 1.19 (0.71-2.01) 0.72

(18.0-58.3) (12.8-’) (20.8-51.9) (15.1-’) (6.7-’)

0.76 1.10 0.91 1.33

(0.34-1.69) (0.55-2.20) (0.43-1.94) (0.57-3.09) 0.48

56 57

31.08 (19.0-56.3) 25.10 (19.9-54.1)

34 32 47

78.9 (19.0-’) 25.6 (18.4-47.9) 23.1 (15.2-37.8)

86 27

29.4 (20.8-58.2) 26.0 (14.7-47.9)

11 38 51 13

’ 47.87 (25.1-89.6) 20.8 (14.7-30.3) 11.2 (5.02-29-44)

30 51 26 6

’ 29.0 (20.0-48.8) 12.85 (7.7-21.2) 11.2 (2.8-’)

23 31 44 15

’ 37.8 (24.3-68.8) 18.4 (12.6-23.1) 20.4 (5.0-30.6)

41 72

18.0 (12.5-26.0) 54.1 (25.6-79.0)

1.18 (0.75-1.86) 0.01 2.03 (1.06-3.91) 2.41 (1.32-4.42) 0.42 1.24 (0.74-2.06) < 0.001 1.36 (0.65-2.81) 2.87 (1.47-5.60) 4.04 (’-’) < 0.001 3.01 (1.49-6.09) 6.90 (3.25-14-65) 8.48 (2.59-27.80) < 0.001 5.48 (1.87-16.09) 11.26 (3.95-32.16) 13.73 (4.31-43-67) < 0.001 0.44 (0.28-0.71)

Tumors were staged at time of diagnosis according to Cancer Staging Manual, AJCC 2002 (6th edition) criteria. Median OS was not reached for T1, N1 and Clinical stage I. The missing values for CI mean that no upper or lower bounds were reached. CI indicates confidence interval; EGFR, epidermal growth factor receptor; HR, hazard ratio; OS, overall survival; R0, negative margin; R1, microscopically positive margin; R2, grossly positive margin.

Francisco, CA),29,32,67,74 MET (SP44) mouse mAb (VentanaRoche, Tucson, AZ),75 rabbit monoclonal EP1454Y (Abcam, Cambridge, MA),76 MET clone 8F11 (Novocastra, Newcastle, UK),65 and MET clone EP1454 (Abcam).77 The mouse mAb MET (SP44) has proven to be correlated with MET gene amplification by SISH.40,66,75 In our study, we chose the mouse mAb MET clone 3D4 (Invitrogen) because it was the most commonly used antibody at the time when we initiated our study,29,32,67,74 and recognizes the cytoplasmic domain of human c-Met. A study showed that, using this clone, antibodies directed against the cytoplasmic domain of c-MET receptor in breast cancer were of higher negative prognostic value then those directed at the membranous domain.78


In our study, coexpression of EGFR and c-MET (12/113; 10%), was strongly associated with poor OS (13 vs. 35 mo for rest of tumors; P = 0.001). Furthermore, when the cohort was subdivided by EGFR status, c-MET positivity only added negative prognostic significance in the EGFR positive subgroup (13 mo in EGFR + /c-MET + vs. 79 mo in EGFR + /cMET , P = 0.07). To our knowledge, we are the first to report this observation. We found only 2 other studies that simultaneously examined both c-MET and EGFR. One study examined 110 patients with high-risk GC that were randomized to platinumbased chemotherapy plus docetaxel with/without radiotherapy.79 In this study which had a primary endpoint of OS, a

Copyright

r






TABLE 3. Multivariate Models of EGFR and c-MET and Selected Tumor Characteristics and OS in Gastric Cancer Patients

EGFR Status Positive Versus Negative (Reference) Multivariate Cox Proportional Hazard Models Markers Location Antrum (reference) GEJ Body T stage 1 (reference) 2 3 4 N stage 0 (reference) 1 2 3 Clinical stage I (reference) II II IV Resection margins R1 or R2 (reference) R0

c-MET Status Positive Versus Negative (Reference)

EGFR/c-MET Positive Versus Rest (Reference)

HR

95% CI

P

HR

95% CI

P

HR

95% CI

P

1.85

0.98-3.49

0.06 0.09

1.20

0.74-1.97

0.46 0.15

2.58

1.21-5.50

0.01 0.15

1.81 2.04

0.92-3.57 1.07-3.86

1.62 1.88

0.82-3.18 0.99-3.55

1.63 1.87

0.83-3.18 0.99-3.53

< 0.001 1.43 5.93 29.30

0.38-5.40 .-. 5.98-144.2

< 0.001 1.39 5.92 28.8

0.37-5.24 .-. 5.89-141.0

0.01 5.81 12.42 31.66

1.26-26.89 2.26-68.26 2.78-61.12

6.43 14.04 34.09

1.41-29.37 2.59-76.29 2.99-388.3

3.81-5.43 .-. 6.07-150.0 0.01

5.52 11.70 33.70

1.19-25.68 2.17-66.25 2.97-382.4

0.23

0.07-2.37 0.01-2.67 0.00-1.49

0.44 0.15 0.03

0.08-2.62 0.01-2.58 0.00-1.35

0.24 0.73

1.43 5.68 30.2 0.008

0.29 0.40 0.16 0.03

< 0.001

0.27 0.47 0.17 0.03

0.08-2.75 0.01-2.83 0.00-1.63

0.18

0.42-1.24

0.70

0.41-1.18

0.43 0.80

0.46-1.39

Tumors were staged at time of diagnosis according to Cancer Staging Manual, AJCC 2002 (6th edition) criteria. P values < 0.05 were considered statistically significant. CI indicates confidence interval; EGFR, epidermal growth factor receptor; GEJ, gastroesophageal junction; HR, hazard ratio; R0, negative margin; R1, microscopically positive margin; R2, grossly positive margin.

TMA was used to assess the prognostic and/or predictive value of several molecular markers including EGFR and c-MET. The receptors showed no prognostic value on their own79 and the study did not specifically report how many of their cases were positive for both EGFR and c-MET, nor did it make any associations for the combination. The second study reported, that from a series of 950 Japanese GC patients, a small minority (5/950) were coexpressing EGFR and c-MET, but it was not reported if this combinationaffected survival.37 They did, however, report that the 2 receptors were interdependent of one another.37 This difference in incidence of double-positive cases may be explained by different antibody use (they used anti-EGFR [3C6] mouse monoclonal antibody [Ventana] and anti-Total c-MET [SP44] rabbit monoclonal antibody [Ventana]), by methodology, and interpretation of what constitutes positivity, and importantly, geographic differences. The fact that 10% of our cohort was EGFR + /c-MET + and had inferior survival, suggests that this group, albeit a minority, may in particular benefit most from targeted inhibition of both the receptors. However, if the receptors are truly interdependent and activation of one receptor upregulates the other via crosstalk,17,35,37,80–82 then tumor reliance on either receptor could also be enough to predict sensitivity to dual inhibition. It has already been proved in non small cell carcinoma cell lines that combined c-MET/EGFR blockade suppresses the emergence of drug-resistant clones, highlighting the tight relationship between c-MET and EGFR.81 In our study, a total 62% (70/113) of GC patients were either positive for EGFR or c-MET or both. The observation that c-MET receptor status had particular significance in EGFR + patients may explain the variability in the reported prognostic value of c-MET. It also provides support for the hypothesis that these 2 Copyright

r

receptors are interdependent and important in tumor genesis, although their relationship is complex. Limitations of our study include the use of TMA blocks instead of entire tumor sections and low patient numbers. However, selecting 4 cores per tumor to ensure representation of histologically heterogeneous areas reduced the impact of TMA use. Furthermore, GC is relatively rare in North America and it is difficult to accrue a large patient series from a single institution. In addition to geographic differences, treatment protocols, tissue processing, surgical approach, and staging protocols vary from region to region, which may affect the prognostic ability of a given receptor. Also, ethnicity, race, and sociodemographic information are not routinely collected in the Canadian medical institutions, although this might change in the future with a strong educational campaign,83 at least in some provinces.84 Strengths of the study include exceptionally long followup and prognostic information based on a representatively treated cohort of Canadian GC patients. To conclude, this study describes the prevalence and prognostic value of EGFR and c-MET expression in a Canadian population of patients undergoing curative intent surgical resection for GC. Both c-MET and EGFR status trended toward poor OS, but only coexpression of EGFR and c-MET defined a subset of GC which showed a statistically significant worse prognosis. To the best of our knowledge, we are the first to report that c-MET activation conferred significant negative impact on EGFR + GC. This finding has important clinical implications, as it may be the basis of developing combination treatment targeted strategies with EGFR and c-MET TKIs. Therefore, assessment in clinical practice of EGFR and c-MET in GC may be warranted in the future, should larger clinical studies confirm our findings.



7



Paliga et al

REFERENCES 1. Hartgrink HH, Jansen EP, Van Grieken NC, et al. Gastric cancer. Lancet. 2009;374:477–490. 2. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin. 2011;61:69–90. 3. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136:E359–E386. 4. Cunningham D, Starling N, Rao S, et al. Capecitabine and oxaliplatin for advanced esophagogastric cancer. N Engl J Med. 2008;358:36–46. 5. Van Cutsem E, Moiseyenko VM, Tjulandin S, et al. Phase III study of docetaxel and cisplatin plus fluorouracil compared with cisplatin and fluorouracil as first-line therapy for advanced gastric cancer: a report of the V325 Study Group. J Clin Oncol. 2006;24: 4991–4997. 6. Wagner AD, Grothe W, Haerting J, et al. Chemotherapy in advanced gastric cancer: a systematic review and meta-analysis based on aggregate data. J Clin Oncol. 2006;24:2903–2909. 7. Wong H, Yau T. Molecular targeted therapies in advanced gastric cancer: does tumor histology matter? Ther Adv Gastroenterol. 2013;6:15–31. 8. Yap TA, Olmos D, Brunetto AT, et al. Phase I trial of a selective c-MET inhibitor ARQ 197 incorporating proof of mechanism pharmacodynamic studies. J Clin Oncol. 2011;29:1271–1279. 9. Herbst RS. Review of epidermal growth factor receptor biology. Int J Radiat Oncol Biol Phys. 2004;59(suppl):21–26. 10. Nicholson RI, Gee JM, Harper ME. EGFR and cancer prognosis. Eur J Cancer (Oxford, England: 1990). 2001;37(suppl 4):S9–S15. 11. Galizia G, Lieto E, Orditura M, et al. Epidermal growth factor receptor (EGFR) expression is associated with a worse prognosis in gastric cancer patients undergoing curative surgery. World J Surg. 2007;31:1458–1468. 12. Lieto E, Ferraraccio F, Orditura M, et al. Expression of vascular endothelial growth factor (VEGF) and epidermal growth factor receptor (EGFR) is an independent prognostic indicator of worse outcome in gastric cancer patients. Ann Surg Oncol. 2008;15: 69–79. 13. Gamboa-Dominguez A, Dominguez-Fonseca C, Quintanilla-Martinez L, et al. Epidermal growth factor receptor expression correlates with poor survival in gastric adenocarcinoma from Mexican patients: a multivariate analysis using a standardized immunohistochemical detection system. Mod Pathol. 2004;17: 579–587. 14. Song HS, Do YR, Kim IH, et al. Prognostic significance of immunohistochemical expression of EGFR and C-erbB-2 oncoprotein in curatively resected gastric cancer. Cancer Res Treat. 2004;36:240–245. 15. Kim MA, Lee HS, Lee HE, et al. EGFR in gastric carcinomas: prognostic significance of protein overexpression and high gene copy number. Histopathology. 2008;52:738–746. 16. Smolen GA, Sordella R, Muir B, et al. Amplification of MET may identify a subset of cancers with extreme sensitivity to the selective tyrosine kinase inhibitor PHA-665752. Proc Natl Acad Sci USA. 2006;103:2316–2321. 17. Sierra JR, Tsao MS. c-MET as a potential therapeutic target and biomarker in cancer. Ther Adv Med Oncol. 2011;3(suppl): S21–S35. 18. Derksen PW, De Gorter DJ, Meijer HP, et al. The hepatocyte growth factor/Met pathway controls proliferation and apoptosis in multiple myeloma. Leukemia. 2003;17:764–774. 19. Xiao GH, Jeffers M, Bellacosa A, et al. Anti-apoptotic signaling by hepatocyte growth factor/Met via the phosphatidylinositol 3kinase/Akt and mitogen-activated protein kinase pathways. Proc Natl Acad Sci USA. 2001;98:247–252. 20. Migliore C, Giordano S. Molecular cancer therapy: can our expectation be MET? Eur J Cancer (Oxford, England: 1990). 2008;44:641–651. 21. Blumenschein GR Jr, Mills GB, Gonzalez-Angulo AM. Targeting the hepatocyte growth factor-cMET axis in cancer therapy. J Clin Oncol. 2012;30:3287–3296.



22. Janjigian YY, Tang LH, Coit DG, et al. MET expression and amplification in patients with localized gastric cancer. Cancer Epidemiol Biomarkers Prev. 2011;20:1021–1027. 23. Kuniyasu H, Yasui W, Kitadai Y, et al. Frequent amplification of the c-met gene in scirrhous type stomach cancer. Biochem Biophys Res Commun. 1992;189:227–232. 24. Seruca R, Suijkerbuijk RF, Gartner F, et al. Increasing levels of MYC and MET co-amplification during tumor progression of a case of gastric cancer. Cancer Genet Cytogenet. 1995;82:140–145. 25. Tsujimoto H, Sugihara H, Hagiwara A, et al. Amplification of growth factor receptor genes and DNA ploidy pattern in the progression of gastric cancer. Virchows Archiv. 1997;431: 383–389. 26. Nessling M, Solinas-Toldo S, Wilgenbus KK, et al. Mapping of chromosomal imbalances in gastric adenocarcinoma revealed amplified protooncogenes MYCN, MET, WNT2, and ERBB2. Genes Chromosomes Cancer. 1998;23:307–316. 27. Sakakura C, Mori T, Sakabe T, et al. Gains, losses, and amplifications of genomic materials in primary gastric cancers analyzed by comparative genomic hybridization. Genes Chromosomes Cancer. 1999;24:299–305. 28. Acunzo M, Romano G, Palmieri D, et al. Cross-talk between MET and EGFR in non-small cell lung cancer involves miR-27a and Sprouty2. Proc Natl Acad Sci USA. 2013;110:8573–8578. 29. Huang TJ, Wang JY, Lin SR, et al. Overexpression of the c-met protooncogene in human gastric carcinoma–correlation to clinical features. Acta Oncol (Stockholm, Sweden). 2001;40:638–643. 30. Ge SH, Wu XJ, Wang XH, et al. Over-expression of metastasisassociated in colon cancer-1 (MACC1) associates with better prognosis of gastric cancer patients. Chin J Cancer Res. 2011;23: 153–159. 31. Ma J, Ma J, Meng Q, et al. Prognostic value and clinical pathology of MACC-1 and c-MET expression in gastric carcinoma. Pathol Oncol Res. 2013;19:821–832. 32. Nakajima M, Sawada H, Yamada Y, et al. The prognostic significance of amplification and overexpression of c-met and cerb B-2 in human gastric carcinomas. Cancer. 1999;85:1894–1902. 33. Qi J, McTigue MA, Rogers A, et al. Multiple mutations and bypass mechanisms can contribute to development of acquired resistance to MET inhibitors. Cancer Res. 2011;71:1081–1091. 34. Yu S, Yu Y, Zhao N, et al. C-Met as a prognostic marker in gastric cancer: a systematic review and meta-analysis. PloS One. 2013; 8:e79137. 35. Guo A, Villen J, Kornhauser J, et al. Signaling networks assembled by oncogenic EGFR and c-Met. Proc Natl Acad Sci USA. 2008;105:692–697. 36. Shida D, Kitayama J, Yamaguchi H, et al. Sphingosine 1phosphate transactivates c-Met as well as epidermal growth factor receptor (EGFR) in human gastric cancer cells. FEBS Lett. 2004;577:333–338. 37. Nagatsuma AK, Aizawa M, Kuwata T, et al. Expression profiles of HER2, EGFR, MET and FGFR2 in a large cohort of patients with gastric adenocarcinoma. Gastric Cancer. 2014;18:227–238. 38. Bachleitner-Hofmann T, Sun MY, Chen CT, et al. HER kinase activation confers resistance to MET tyrosine kinase inhibition in MET oncogene-addicted gastric cancer cells. Mol Cancer Ther. 2008;7:3499–3508. 39. Corso S, Ghiso E, Cepero V, et al. Activation of HER family members in gastric carcinoma cells mediates resistance to MET inhibition. Mol Cancer. 2010;9:121. 40. Engelman JA, Zejnullahu K, Mitsudomi T, et al. MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science (New York, NY). 2007;316:1039–1043. 41. Karamouzis MV, Konstantinopoulos PA, Papavassiliou AG. Targeting MET as a strategy to overcome crosstalk-related resistance to EGFR inhibitors. Lancet Oncol. 2009;10:709–717. 42. Yamada T, Takeuchi S, Kita K, et al. Hepatocyte growth factor induces resistance to anti-epidermal growth factor receptor antibody in lung cancer. J Thorac Oncol. 2012;7:272–280. 43. Puri N, Salgia R. Synergism of EGFR and c-Met pathways, crosstalk and inhibition, in non-small cell lung cancer. J Carcinog. 2008;7:9.

Copyright

r





44. Xu L, Nilsson MB, Saintigny P, et al. Epidermal growth factor receptor regulates MET levels and invasiveness through hypoxiainducible factor-1alpha in non-small cell lung cancer cells. Oncogene. 2010;29:2616–2627. 45. Castoldi R, Ecker V, Wiehle L, et al. A novel bispecific EGFR/ Met antibody blocks tumor-promoting phenotypic effects induced by resistance to EGFR inhibition and has potent antitumor activity. Oncogene. 2013;32:5593–5601. 46. Bang YJ, Van Cutsem E, Feyereislova A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376:687–697. 47. Brevet M, Arcila M, Ladanyi M. Assessment of EGFR mutation status in lung adenocarcinoma by immunohistochemistry using antibodies specific to the two major forms of mutant EGFR. J Mol Diagn. 2010;12:169–176. 48. Hasanovic A, Ang D, Moreira AL, et al. Use of mutation specific antibodies to detect EGFR status in small biopsy and cytology specimens of lung adenocarcinoma. Lung Cancer (Amsterdam, The Netherlands). 2012;77:299–305. 49. Schemper M, Smith TL. A note on quantifying follow-up in studies of failure time. Control Clin Trials. 1996;17:343–346. 50. McShane LM, Altman DG, Sauerbrei W, et al. Reporting recommendations for tumor marker prognostic studies (REMARK). J Natl Cancer Inst. 2005;97:1180–1184. 51. Maldonado G, Greenland S. Simulation study of confounderselection strategies. Am J Epidemiol. 1993;138:923–936. 52. Harrell FE Jr, Califf RM, Pryor DB, et al. EValuating the yield of medical tests. JAMA. 1982;247:2543–2546. 53. Harrell FE Jr, Lee KL, Califf RM, et al. Regression modelling strategies for improved prognostic prediction. Stat Med. 1984;3: 143–152. 54. Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika. 1994;81:515–526. 55. Begnami MD, Fukuda E, Fregnani JH, et al. Prognostic implications of altered human epidermal growth factor receptors (HERs) in gastric carcinomas: HER2 and HER3 are predictors of poor outcome. J Clin Oncol. 2011;29:3030–3036. 56. Kimura M, Tsuda H, Morita D, et al. A proposal for diagnostically meaningful criteria to classify increased epidermal growth factor receptor and c-erbB-2 gene copy numbers in gastric carcinoma, based on correlation of fluorescence in situ hybridization and immunohistochemical measurements. Virchows Archiv. 2004;445:255–262. 57. Kimura M, Tsuda H, Morita D, et al. Usefulness and limitation of multiple endoscopic biopsy sampling for epidermal growth factor receptor and c-erbB-2 testing in patients with gastric adenocarcinoma. Jpn J Clin Oncol. 2005;35:324–331. 58. Atmaca A, Werner D, Pauligk C, et al. The prognostic impact of epidermal growth factor receptor in patients with metastatic gastric cancer. BMC Cancer. 2012;12:524. 59. Mokhtari M, Ardestani MM, Movahedipour M. An immunohistochemical study of EGFR expression in colorectal cancer and its correlation with lymph nodes status and tumor grade. J Res Med Sci. 2012;17:741–744. 60. Terashima M, Kitada K, Ochiai A, et al. Impact of expression of human epidermal growth factor receptors EGFR and ERBB2 on survival in stage II/III gastric cancer. Clin Cancer Res. 2012;18:5992–6000. 61. Oh HS, Eom DW, Kang GH, et al. Prognostic implications of EGFR and HER-2 alteration assessed by immunohistochemistry and silver in situ hybridization in gastric cancer patients following curative resection. Gastric Cancer. 2013;17:402–411. 62. Al-Moundhri MS, Al-Hadabi I, Al-Mawaly K, et al. Prognostic significance of cyclooxygenase-2, epidermal growth factor receptor 1, and microvascular density in gastric cancer. Med Oncol (Northwood, London, England). 2012;29:1739–1747. 63. Kim JH, Kim MA, Lee HS, et al. Comparative analysis of protein expressions in primary and metastatic gastric carcinomas. Hun Pathol. 2009;40:314–322. 64. An X, Wang F, Shao Q, et al. MET amplification is not rare and predicts unfavorable clinical outcomes in patients with recurrent/metastatic gastric cancer after chemotherapy. Cancer. 2014;120:675–682.

Copyright

r


65. Drebber U, Baldus SE, Nolden B, et al. The overexpression of c-met as a prognostic indicator for gastric carcinoma compared to p53 and p21 nuclear accumulation. Oncol Rep. 2008;19:1477–1483. 66. Lee HE, Kim MA, Lee HS, et al. MET in gastric carcinomas: comparison between protein expression and gene copy number and impact on clinical outcome. Br J Cancer. 2012;107:325–333. 67. Taniguchi K, Yonemura Y, Nojima N, et al. The relation between the growth patterns of gastric carcinoma and the expression of hepatocyte growth factor receptor (c-met), autocrine motility factor receptor, and urokinase-type plasminogen activator receptor. Cancer. 1998;82:2112–2122. 68. Ha SY, Lee J, Kang SY, et al. MET overexpression assessed by new interpretation method predicts gene amplification and poor survival in advanced gastric carcinomas. Mod Pathol. 2013;26: 1632–1641. 69. Li Y, Chen CQ, He YL, et al. Abnormal expression of E-cadherin in tumor cells is associated with poor prognosis of gastric carcinoma. J Surg Oncol. 2012;106:304–310. 70. Peng Z, Zhu Y, Wang Q, et al. Prognostic significance of MET amplification and expression in gastric cancer: a systematic review with meta-analysis. PloS One. 2014;9:e84502. 71. Bhargava R, Chen B, Klimstra DS, et al. Comparison of two antibodies for immunohistochemical evaluation of epidermal growth factor receptor expression in colorectal carcinomas, adenomas, and normal mucosa. Cancer. 2006;106:1857–1862. 72. Lee WY, Chen HH, Chow NH, et al. Prognostic significance of co-expression of RON and MET receptors in node-negative breast cancer patients. Clin Cancer Res. 2005;11:2222–2228. 73. Cheng HL, Trink B, Tzai TS, et al. Overexpression of c-met as a prognostic indicator for transitional cell carcinoma of the urinary bladder: a comparison with p53 nuclear accumulation. J Clin Oncol. 2002;20:1544–1550. 74. Han SU, Lee HY, Lee JH, et al. Modulation of E-cadherin by hepatocyte growth factor induces aggressiveness of gastric carcinoma. Ann Surg. 2005;242:676–683. 75. Arriola E, Canadas I, Arumi-Uria M, et al. 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. 2011;105:814–823. 76. Toiyama Y, Yasuda H, Saigusa S, et al. Co-expression of hepatocyte growth factor and c-Met predicts peritoneal dissemination established by autocrine hepatocyte growth factor/c-Met signaling in gastric cancer. Int J Cancer. 2012;130:2912–2921. 77. Jacobsen F, Ashtiani SN, Tennstedt P, et al. High c-MET expression is frequent but not associated with early PSA recurrence in prostate cancer. Exp Ther Med. 2013;5:102–106. 78. Kang JY, Dolled-Filhart M, Ocal IT, et al. Tissue microarray analysis of hepatocyte growth factor/Met pathway components reveals a role for Met, matriptase, and hepatocyte growth factor activator inhibitor 1 in the progression of node-negative breast cancer. Cancer Res. 2003;63:1101–1105. 79. Bamias A, Karina M, Papakostas P, et al. A randomized phase III study of adjuvant platinum/docetaxel chemotherapy with or without radiation therapy in patients with gastric cancer. Cancer Chemother Pharmacol. 2010;65:1009–1021. 80. Matsubara D, Ishikawa S, Sachiko O, et al. Co-activation of epidermal growth factor receptor and c-MET defines a distinct subset of lung adenocarcinomas. Am J Pathol. 2010;177:2191–2204. 81. McDermott U, Pusapati RV, Christensen JG, et al. Acquired resistance of non-small cell lung cancer cells to MET kinase inhibition is mediated by a switch to epidermal growth factor receptor dependency. Cancer Res. 2010;70:1625–1634. 82. Agarwal S, Zerillo C, Kolmakova J, et al. Association of constitutively activated hepatocyte growth factor receptor (Met) with resistance to a dual EGFR/Her2 inhibitor in non-small-cell lung cancer cells. Br J Cancer. 2009;100:941–949. 83. Kirst M, Shankardass K, Bomze S, et al. Sociodemographic data collection for health equity measurement: a mixed methods study examining public opinions. Int J Equity Health. 2013;12:75–75. 84. Quan H, Wong A, Johnson D, et al. The public endorses collection of ethnicity information in hospital: implications for routine data capture in Canadian Health Systems. Healthc Policy. 2006;1:55–64.



9


Assessment and prognostic analysis of EGFR, HER2, and HER3 protein expression in surgically resected gastric adenocarcinomas.

Clinicopathologic characteristics and prognostic significance of EGFR and p53 mutations in surgically resected lung adenocarcinomas ≤2 cm in maximal dimension.

EGFR family and cMet expression profiles and prognostic significance in esophagogastric adenocarcinoma.

FGFR2, HER2 and cMet in gastric adenocarcinoma: detection, prognostic significance and assessment of downstream pathway activation.

Significance of NQO1 overexpression for prognostic evaluation of gastric adenocarcinoma.

Prognostic significance of overexpression of p53 in uterine endometrioid adenocarcinomas with an analysis of nuclear grade.

Clinicopathological and prognostic significance of HER2 overexpression in gastric cancer: a meta-analysis of the literature.

The predictive and prognostic values of factors associated with visceral pleural involvement in resected lung adenocarcinomas.

Clinicopathological and prognostic significance of CD24 overexpression in patients with gastric cancer: a meta-analysis.

Clinicopathological correlation and prognostic significance of sonic hedgehog protein overexpression in human gastric cancer.

Clinical and CT characteristics of surgically resected lung adenocarcinomas harboring ALK rearrangements or EGFR mutations.

Prognostic significance of survivin in resected gallbladder cancer.

Prognostic significance of computed tomography in resected N2 lung cancer.

The Association between EGFR and cMET Expression and Phosphorylation and Its Prognostic Implication in Patients with Breast Cancer.

COX-2 overexpression in resected pancreatic head adenocarcinomas correlates with favourable prognosis.

The significance of LRPPRC overexpression in gastric cancer.

Prognostic significance of Tspan9 in gastric cancer.

Prognostic significance of TMPRSS4 in gastric cancer.

Molecular, morphological and survival analysis of 177 resected pancreatic ductal adenocarcinomas (PDACs): Identification of prognostic subtypes.

Prognostic significance of the co-overexpression of fibroblast growth factor receptors 1, 2 and 4 in gastric cancer.

Prognostic value of epidermal growth factor receptor mutations in resected lung adenocarcinomas.

Prognostic and therapeutic implications of aromatase expression in lung adenocarcinomas with EGFR mutations.

Prognostic significance of CD168 overexpression in colorectal cancer.

Prognostic significance of microRNA expression in completely resected lung adenocarcinoma and the associated response to erlotinib.