A study of indefinite for dysplasia in Barrett's oesophagus: reproducibility of diagnosis, clinical outcomes and predicting progression with AMACR (alpha-methylacyl-CoA-racemase).

Aims: To assess interobserver variation in the diagnosis of dysplasia in Barrett's oesophagus, especially indefinite dysplasia (IND) using the revised Vienna classification. A secondary aim was to study clinical outcome of IND cases and to evaluate expression of alpha-methyl-CoA racemase (AMACR) as a marker predictive of progression.

Methods And Results: Cases of Barrett's oesophagus and dysplasia over a 20 year period were assessed. Three experienced histopathologists reviewed 101 cases on set criteria in a blinded fashion. Slides were immunostained for AMACR and evaluated for the presence, extent and location of AMACR expression. Clinical and progression data were collected. Overall agreement for the diagnosis of dysplasia was fair (k = 0.35) but that for IND was poor (k = 0.18). 6 IND cases progressed after a median follow-up of 31.4 months to a higher grade. The sensitivity of AMACR for the detection of abnormality was 22% for IND and specificity 100%. The positive predictive value of AMACR for progression was 0.44 and the negative predictive value was 0.92.

Conclusion: Fair agreement was achieved for the diagnosis of dysplasia but poor agreement for IND. A proportion of IND cases progress. Re-diagnosis or consensus diagnosis did not predict progression. AMACR shows promise as a marker to indicate IND patients in need of more intensive surveillance.