Introduction Genetic factors play a crucial role in the predisposition to Barrett’s epithelium (BE) and progression to oesophageal adenocarcinoma (OA). Using next generation sequencing (NGS) methods on frozen tissue, TP53 mutations have been identified in a stage specific manor in BE (1). The standard method of preserving BE is formalin fixing and paraffin embedding (FFPE). NGS on tiny FFPE specimens is challenging. We examined the prevalence of TP53 genomic aberrations and p53 expression in tiny FFPE BE biopsies using an adapted NGS platform.
Method 37 cases including non-dysplastic BE (NDBE; n = 7), low grade dysplasia (LGD; n = 7), high grade dysplasia (HGD; n = 9), intramucosal cancer (IMC; n = 6) and invasive OA (IOA; n = 8) were sequenced and immunostained. We used an inhouse, high throughput multiplexing method for targeted mutational screening using NGS on DNA extracted from FFPE tissue and analysed the files with our unique hidden Markov analytical method. Automated p53 IHC using clone D07 was used to correlate detected TP53 mutations with tissue expression.
Results DNA extracted varied from 2–50 ng, including some highly degraded DNA material. TP53 was mutated in 40.5% (15/37) of cases; 28.6% of NDBE (2/7), 71.4% of LGD (5/7), 44.4% of HGD (4/9), 33.3% of IMC (2/6) and 25% of IOA (2/8). Interestingly, only 40% (6/15) of cases showed both TP53 mutations and p53 aberrations (over- or absence of expression), 60% (9/15) had mutations but no protein aberration, and 45% (5/11) had p53 aberration but no mutational abnormality. The data quality was excellent with all cases having a depth of coverage of at least x800 (mean coverage x1200 (Figure 1A). The level of p53 staining did not however correlate with mutation data (Figure 1B).
Conclusion These results show discordance between detection of TP53 using mutation analysis and current p53 IHC methods. Furthermore, we identified patients with TP53 mutations in NDBE, which may pre-empt progression to OA. Identifying early genomic aberrations driving the development of OA from BE holds the key to accurately risk stratifying BE patients into more aggressive treatment strategies or frequent surveillance regimes.
Disclosure of interest None Declared.
Weaver et al. Nat Genet. 2014;46(8):837-43