Tumor Biol. (2015) 36:4993–5000 DOI 10.1007/s13277-015-3150-7
RESEARCH ARTICLE
MET is a predictive factor for late recurrence but not for overall survival of early stage hepatocellular carcinoma Young Wha Koh & Yang-Soon Park & Hyo Jeong Kang & Ju Hyun Shim & Eunsil Yu
Received: 3 September 2014 / Accepted: 26 January 2015 / Published online: 10 February 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015
Abstract The MET and RON receptors are tyrosine kinases that form a non-covalent complex on the cell surface that functions in several steps of tumor progression. The purpose of this study was to determine the clinical significance of MET and RON expression on long-term survival and recurrence after curative resection in a large cohort of hepatocellular carcinoma (HCC) patients. We performed immunohistochemical analyses on microarrays of the tumors using antibodies against MET and RON. We evaluated the prognostic value of biomarker expression using Cox regression and the Kaplan– Meier method in 490 HCC patients. MET-positive patients had higher overall recurrence rates than MET-negative patients (P=0.041); however, MET positivity was not associated with overall survival (OS) (P=0.249). RON was not associated with overall recurrence rates and OS. MET was independently associated with late but not early phase recurrence. Particularly, the prognostic significance of MET is limited in early stage disease. MET+/RON+ patients had higher overall recurrence rates than those with the other expression patterns (P=0.071), although the result did not reach statistical significance. Immunohistological activation of MET expression has Electronic supplementary material The online version of this article (doi:10.1007/s13277-015-3150-7) contains supplementary material, which is available to authorized users. Y. W. Koh Department of Pathology, Ajou University School of Medicine, Suwon, South Korea Y.70 % of one of the two core cylinders from each patient. We also used a matched normal hepatic tissue to compare the intensity between normal hepatic tissue and hepatocellular carcinoma. Normal hepatic tissue was negative or weakly reactive for MET and RON in Supplementary Fig. 1. All of positive HCC cells stained much stronger for MET or RON than normal hepatic tissue. RON and MET overexpression in each tumor was examined using a fourtier scoring system based on the literature [12, 20] as follows: 0 = no reactivity in invasive tumor cells; 1+ = weak reactivity, faint or light membrane, or cytoplasmic staining in any proportion of the tumor cells; 2+ = moderate reactivity, intermediate membrane, or cytoplasmic staining between 1+ and 3+ in at least 30 % of tumor cells; and 3+ = strong reactivity, intense complete membrane, or cytoplasmic staining in more than 30 % of tumor cells. We defined scores 0 and 1+ as negative and scores 2+ and 3+ as positive. All interpretations of MET and RON staining patterns were performed by a colleague blinded to the clinical outcomes. Statistical analysis
Histopathological analysis and immunohistochemistry Paraffin-embedded tumor tissues were available for all included patients.
We used the Kaplan–Meier method and log rank comparisons to plot cumulative probability. Multivariate prognostic analyses were performed using Cox proportional hazards regression
Tumor Biol. (2015) 36:4993–5000
4995
model. Categorical variables were compared using the chisquared two-sided Pearson’s test, Fisher’s test, and linear-bylinear association test. The cutoff time for early recurrence was set at 2 years [24]. All statistical analyses were performed by the SPSS statistical software program (version 18.0; SPSS, Chicago, IL). P50 years Male gender ES nuclear grade 1 or 2 3 or 4 HBV present HCV present Size >5 cm Solitary tumor pT stage pT1 pT2 or pT3
264 (53.9) 400 (81.6)
Liver cirrhosis present Serum AFP >100 ng/mL Microvascular invasion Glisson capsule invasion
289 (59 %) 233 (48 %) 152 (31 %) 52 (10.6 %)
147 (30 %) 343 (70 %) 422 (86.1 %) 39 (8 %) 183 (37.3 %) 455 (92.9 %) 316 (64.5 %) 174 (35.5 %)
AFP alpha-fetoprotein, ES nuclear grade, Edmondson-Steiner nuclear grade, HBV hepatitis B virus, HCV hepatitis C virus
and 209 (42.7 %) cases were negative for both MET and RON. MET expression was significantly correlated with RON expression (P1 pT1 vs. pT2 or 3 (−) vs. (+)
1.178 1.216 1.275 1.610
0.94–1.47 0.80–1.84 1.02–1.59 1.28–2.02
0.148 0.356 0.033 100 (−) vs. (+) (−) vs. (+) (−) vs. (+) (−) vs. (+) Other vs. MET(+) and RON(+)
0.869 1.302 1.541 1.252 1.138 1.272
0.69–1.08 1.03–1.63 1.11–2.14 1.01–1.55 0.91–1.42 0.97–1.65
0.205 0.024 0.01 0.044 0.253 0.074
0.898 1.524 1.819 1.180 0.910 0.958
0.68–1.18 1.14–2.03 1.23–2.68 0.89–1.57 0.68–1.21 0.67–1.37
0.452 0.004 0.003 0.251 0.525 0.816
1–2 vs. 3–4 pT1 vs. pT2 or 3 (−) vs. (+) (−) vs. (+) (−) vs. (+)
HR – 1.212 1.613 1.549 1.250
95 % CI – 0.96–1.52 1.28–2.02 1.11–2.16 1.01–1.55
P value – 0.097
RESEARCH ARTICLE
MET is a predictive factor for late recurrence but not for overall survival of early stage hepatocellular carcinoma Young Wha Koh & Yang-Soon Park & Hyo Jeong Kang & Ju Hyun Shim & Eunsil Yu
Received: 3 September 2014 / Accepted: 26 January 2015 / Published online: 10 February 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015
Abstract The MET and RON receptors are tyrosine kinases that form a non-covalent complex on the cell surface that functions in several steps of tumor progression. The purpose of this study was to determine the clinical significance of MET and RON expression on long-term survival and recurrence after curative resection in a large cohort of hepatocellular carcinoma (HCC) patients. We performed immunohistochemical analyses on microarrays of the tumors using antibodies against MET and RON. We evaluated the prognostic value of biomarker expression using Cox regression and the Kaplan– Meier method in 490 HCC patients. MET-positive patients had higher overall recurrence rates than MET-negative patients (P=0.041); however, MET positivity was not associated with overall survival (OS) (P=0.249). RON was not associated with overall recurrence rates and OS. MET was independently associated with late but not early phase recurrence. Particularly, the prognostic significance of MET is limited in early stage disease. MET+/RON+ patients had higher overall recurrence rates than those with the other expression patterns (P=0.071), although the result did not reach statistical significance. Immunohistological activation of MET expression has Electronic supplementary material The online version of this article (doi:10.1007/s13277-015-3150-7) contains supplementary material, which is available to authorized users. Y. W. Koh Department of Pathology, Ajou University School of Medicine, Suwon, South Korea Y.70 % of one of the two core cylinders from each patient. We also used a matched normal hepatic tissue to compare the intensity between normal hepatic tissue and hepatocellular carcinoma. Normal hepatic tissue was negative or weakly reactive for MET and RON in Supplementary Fig. 1. All of positive HCC cells stained much stronger for MET or RON than normal hepatic tissue. RON and MET overexpression in each tumor was examined using a fourtier scoring system based on the literature [12, 20] as follows: 0 = no reactivity in invasive tumor cells; 1+ = weak reactivity, faint or light membrane, or cytoplasmic staining in any proportion of the tumor cells; 2+ = moderate reactivity, intermediate membrane, or cytoplasmic staining between 1+ and 3+ in at least 30 % of tumor cells; and 3+ = strong reactivity, intense complete membrane, or cytoplasmic staining in more than 30 % of tumor cells. We defined scores 0 and 1+ as negative and scores 2+ and 3+ as positive. All interpretations of MET and RON staining patterns were performed by a colleague blinded to the clinical outcomes. Statistical analysis
Histopathological analysis and immunohistochemistry Paraffin-embedded tumor tissues were available for all included patients.
We used the Kaplan–Meier method and log rank comparisons to plot cumulative probability. Multivariate prognostic analyses were performed using Cox proportional hazards regression
Tumor Biol. (2015) 36:4993–5000
4995
model. Categorical variables were compared using the chisquared two-sided Pearson’s test, Fisher’s test, and linear-bylinear association test. The cutoff time for early recurrence was set at 2 years [24]. All statistical analyses were performed by the SPSS statistical software program (version 18.0; SPSS, Chicago, IL). P50 years Male gender ES nuclear grade 1 or 2 3 or 4 HBV present HCV present Size >5 cm Solitary tumor pT stage pT1 pT2 or pT3
264 (53.9) 400 (81.6)
Liver cirrhosis present Serum AFP >100 ng/mL Microvascular invasion Glisson capsule invasion
289 (59 %) 233 (48 %) 152 (31 %) 52 (10.6 %)
147 (30 %) 343 (70 %) 422 (86.1 %) 39 (8 %) 183 (37.3 %) 455 (92.9 %) 316 (64.5 %) 174 (35.5 %)
AFP alpha-fetoprotein, ES nuclear grade, Edmondson-Steiner nuclear grade, HBV hepatitis B virus, HCV hepatitis C virus
and 209 (42.7 %) cases were negative for both MET and RON. MET expression was significantly correlated with RON expression (P1 pT1 vs. pT2 or 3 (−) vs. (+)
1.178 1.216 1.275 1.610
0.94–1.47 0.80–1.84 1.02–1.59 1.28–2.02
0.148 0.356 0.033 100 (−) vs. (+) (−) vs. (+) (−) vs. (+) (−) vs. (+) Other vs. MET(+) and RON(+)
0.869 1.302 1.541 1.252 1.138 1.272
0.69–1.08 1.03–1.63 1.11–2.14 1.01–1.55 0.91–1.42 0.97–1.65
0.205 0.024 0.01 0.044 0.253 0.074
0.898 1.524 1.819 1.180 0.910 0.958
0.68–1.18 1.14–2.03 1.23–2.68 0.89–1.57 0.68–1.21 0.67–1.37
0.452 0.004 0.003 0.251 0.525 0.816
1–2 vs. 3–4 pT1 vs. pT2 or 3 (−) vs. (+) (−) vs. (+) (−) vs. (+)
HR – 1.212 1.613 1.549 1.250
95 % CI – 0.96–1.52 1.28–2.02 1.11–2.16 1.01–1.55
P value – 0.097
