Background: Evidence suggests that MET amplification and/or Met expression are biomarkers of poor prognosis for gastroesophageal cancer (GEC). We previously reported both MET gene copy and amplification by fluorescence in situ hybridization (FISH) and Met protein expression by immunohistochemistry (IHC) as negative prognostic biomarkers for survival in univariate analyses in a large cohort of GEC samples. Here we contrasted these results with Met expression determined by selected reaction monitoring mass spectrometry (SRM-MS), in order to evaluate the distribution and prognostic effect of tumor Met-SRM (along with Her2-SRM) expression after adjusting for other clinical and pathologic covariates. Methods: Formalin-fixed, paraffin-embedded (FFPE) GEC samples (N=394) primarily from early-stage tumors were collected in the United States and Italy. Samples were analyzed by FISH for MET gene amplification (Dako MET/CEN-7 IQFISH Probe Mix, Research Use Only) and by IHC for Met protein expression (Dako MET IHC assay, MET4 antibody, investigational use only), and by Met-SRM (Catenacci et al. PLoS One 2014). Samples with a MET/CEP7 ratio ≥2.0 were considered amplified, whereas samples with ≥25% Met tumor membrane staining by IHC (≥1+ intensity) were considered Met positive. Optimal Met-SRM cut-offs deemed prognostic of survival were identified using the empirical value corresponding to the most significant p-values. Spearman's rank correlation coefficient was used to assess correlations between parameters. Cox proportional hazards models and Kaplan–Meier estimates were applied to explore relationships between MET status (by FISH, IHC, and SRM) and overall survival, and other clinical characteristics. Results: Of the 282 evaluable Met-SRM samples, the mean Met-SRM level was 101.9amol/ug, and ranged <150–3245.5amol/ug, with 18% having detectable levels (>200amol/ug). Using an identified cut-off of 400amol/ug, 22 (7.8%) of the samples were considered Met-SRM+. MET gene amplification (MET/CEP7 ≥2.0) was observed in 16 of 344 samples (4.7%). Of the 332 evaluable IHC samples, 117 (35.2%) were positive for Met protein expression. Although SRM, FISH, and IHC were all prognostic of survival in univariate analyses at their respective cut-offs, after adjusting for covariates only MET gene amplification (HR 2.55, p=0.008), and Met-SRM (HR 1.85, p=0.038) remained independently prognostic. A correlation of Met-SRM with Her2-SRM was observed (p=0.0127), and Her2-SRM expression levels were independently prognostic after adjusting for covariates. Conclusions: Met-SRM ‘high’ expression was observed in 7.8% and MET amplification in ∼5% of this large set of GEC samples, and both were independently observed to be negative prognostic biomarkers. These results indicate that MET gene amplification and Met-SRM protein expression are prognostic of poor outcomes in GEC. Conflict of interest: Advisory Board: Amgen. Oncoplex Dx. Corporate-sponsored Research: Amgen. Oncoplex Dx.
Evaluation of MET using FISH, IHC, or mass spectrometry as a prognostic biomarker in patients with gastroesophageal cancer
RUZZO, ANNAMARIA;
2015
Abstract
Background: Evidence suggests that MET amplification and/or Met expression are biomarkers of poor prognosis for gastroesophageal cancer (GEC). We previously reported both MET gene copy and amplification by fluorescence in situ hybridization (FISH) and Met protein expression by immunohistochemistry (IHC) as negative prognostic biomarkers for survival in univariate analyses in a large cohort of GEC samples. Here we contrasted these results with Met expression determined by selected reaction monitoring mass spectrometry (SRM-MS), in order to evaluate the distribution and prognostic effect of tumor Met-SRM (along with Her2-SRM) expression after adjusting for other clinical and pathologic covariates. Methods: Formalin-fixed, paraffin-embedded (FFPE) GEC samples (N=394) primarily from early-stage tumors were collected in the United States and Italy. Samples were analyzed by FISH for MET gene amplification (Dako MET/CEN-7 IQFISH Probe Mix, Research Use Only) and by IHC for Met protein expression (Dako MET IHC assay, MET4 antibody, investigational use only), and by Met-SRM (Catenacci et al. PLoS One 2014). Samples with a MET/CEP7 ratio ≥2.0 were considered amplified, whereas samples with ≥25% Met tumor membrane staining by IHC (≥1+ intensity) were considered Met positive. Optimal Met-SRM cut-offs deemed prognostic of survival were identified using the empirical value corresponding to the most significant p-values. Spearman's rank correlation coefficient was used to assess correlations between parameters. Cox proportional hazards models and Kaplan–Meier estimates were applied to explore relationships between MET status (by FISH, IHC, and SRM) and overall survival, and other clinical characteristics. Results: Of the 282 evaluable Met-SRM samples, the mean Met-SRM level was 101.9amol/ug, and ranged <150–3245.5amol/ug, with 18% having detectable levels (>200amol/ug). Using an identified cut-off of 400amol/ug, 22 (7.8%) of the samples were considered Met-SRM+. MET gene amplification (MET/CEP7 ≥2.0) was observed in 16 of 344 samples (4.7%). Of the 332 evaluable IHC samples, 117 (35.2%) were positive for Met protein expression. Although SRM, FISH, and IHC were all prognostic of survival in univariate analyses at their respective cut-offs, after adjusting for covariates only MET gene amplification (HR 2.55, p=0.008), and Met-SRM (HR 1.85, p=0.038) remained independently prognostic. A correlation of Met-SRM with Her2-SRM was observed (p=0.0127), and Her2-SRM expression levels were independently prognostic after adjusting for covariates. Conclusions: Met-SRM ‘high’ expression was observed in 7.8% and MET amplification in ∼5% of this large set of GEC samples, and both were independently observed to be negative prognostic biomarkers. These results indicate that MET gene amplification and Met-SRM protein expression are prognostic of poor outcomes in GEC. Conflict of interest: Advisory Board: Amgen. Oncoplex Dx. Corporate-sponsored Research: Amgen. Oncoplex Dx.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
