The clinical impact of hypoxia-regulated gene expression in loco-regional gastroesophageal cancer

Purpose/Objective: In a former study (1), the hypoxia gene expression classifier, developed in head and neck squamous cell carcinomas, was applied in 89 patients with loco-regional gastroesophageal cancer (GC). Analysis of the 15 genes was indicative of hypoxia being more profound in esophagus squamous cell carcinomas (ESCC) compared with adenocarcinomas of the esophago-gastric junction and the stomach (AC), and was a potential prognostic marker in patients with ESCC. The purpose of the present study was to confirm these results. Materials and Methods: The study population consisted of 152 patients with GC treated with neoadjuvant or definitive chemoradiotherapy or perioperative chemotherapy. Based on formalin-fixed, paraffin-embedded, diagnostic biopsies, gene expression of the 15 hypoxia-induced and pH-independent genes was obtained with qPCR. Sufficient amounts of RNA for gene analyses were available in 135 patients; ESCC: 89 patients (66%), AC: 43 patients (32%) and other carcinomas: 3 patients (2%). Comparing the cohorts of the original and present study, patient characteristics were similar except for a significantly lower T- and N-stage in the present cohort. Tumor specimens were ranked for each of the 15 genes and the final rank was the median of the individual gene ranks. A low ranking correlated with a more hypoxic genotype and a high ranking with a less hypoxic genotype. In addition, a gene clustering analysis was performed. Results: Consistent with the original study, an unsupervised hierarchical clustering of the 15 genes showed inter-group heterogeneity between the ESCC and the AC group, and intra-group heterogeneity in the ESCC group. Also consistent with the original study, no intra-group variability was observed for patients with AC. Hence, the analysis suggested that hypoxia was more profound in ESCC. In contrast, the previous indication of a prognostic value of hypoxia in patients with ESCC was not reproducible when dividing patients according to ranking, divided by tertiles and grouping upper tertile versus mid plus lower tertile (Overall survival: HR: 1.01, P=0.98 (Fig. 1)); Disease-specific survival: HR: 1.10, P=0.776). However, in line with the previous study, the hypoxic status did not correlate with treatment response in the ESCC group (P=1.00). Conclusions: Gene expression analysis of the hypoxia classifier confirmed that ESCC hold more hypoxic tumors and display greater heterogeneity compared to AC. However, previous indications that the hypoxia classifier might hold prognostic significance in ESCC patients could not be confirmed. Ongoing work includes in vitro studies of esophageal cancer cell lines in order to identify alternative hypoxia induced genes and to further explore the prognostic value of hypoxia in patients with loco-regional gastroesophageal cancer. (Figure Presented).