Introduction Gastric cancer remains the fourth most prevalent cancer and the second leading cause of cancer-related death in the world. Prognostic models for survival are inadequate. In this retrospective study of archival material we used a panel of 32 RNA probes to characterise gene expression in gastric cancer patients with aggressive disease and those who survived long term.

Methods The University Hospital pathology database was searched for all gastric adenocarcinomas diagnosed between April 2005 and September 2006. The patients' age, sex, tumour stage and survival was recorded. Sixty cancers (n=60) were identified; paraffin sections retrieved and processed using HTG Molecular's qNPA technology to hybridise RNA probes specific to each gene sequence. Tissues sections were placed directly into the 96 plate wells before hybridisation and automated reading. Each plate contained generic anchor sequences hybridised to linker probes, creating gene-specific hybridisation spots to measure each qNPA probe. Each tissue section was analysed 2 or 3 times, gene expression quantitated and an average numerical value derived. Data sets were analysed for normal distribution using the Kolmogorov–Smirnov statistic method. T-test was used with Welch correction on those samples which passed while Mann–Whitney test was used on those that did not. The expression of the 32 genes was compared between patients with metastatic disease (n=30) and 5-year survivors (n=10).

Results When patients with metastatic disease were compared with 5-year survivors significant increased gene expression was noted for heat shock protein (HSP-90), p=0.016 and KRAS, p=0.046. When the pattern of gene expression between those who survived less than a year (n=40) and those surviving 5 years (n=10) was compared then HSP-90 and GHRL were shown also to have altered expression. A variety of genes including TFT1, HSPD1, BCAS1, CAPDH, GHRL were globally enhanced among virtually all samples.

Conclusion HSP-90 a gene encoding a chaperone protein implicated in carcinogenesis exhibited increased expression in metastatic disease. Up-regulation of HSP-90 should help cancer cells adapt to stress conferring a survival advantage. Thus finding relative over-expression in cancers which have progressed to metastatic disease suggests a possible role as a prognostic marker. This study has identified candidate genes that could contribute to a prognostic model for gastric cancer utilising qNPA technology and demonstrates the up-regulation of HSP-90 and KRAS in advanced gastric cancer.

Competing interests None declared.