p53 staining in Barrett's oesophagus was reproducible and resolved over half of indefinite diagnoses

Design

Nationwide consecutive consultation series with interobserver study

Population

1,146 Barrett's oesophagus cases (1,704 biopsy levels) referred to a Dutch expert panel

Primary outcome

Concordance, interobserver agreement and diagnostic reclassification with p53 IHC

Effect

Interobserver kappa 0.65 (95% CI 0.61 to 0.69); indefinite biopsies reclassified to LGD 37.1%, to non-dysplastic 17.0%

The Dutch Esophageal Pathology Panel reviews Barrett's oesophagus biopsies referred from across the Netherlands. This study, published in June, analysed 1,146 consecutive consultation cases (1,704 biopsy levels), each assessed by a median of 12 expert pathologists, on haematoxylin and eosin alone, on p53 immunohistochemistry patterns, and on both together.

Agreement between referring hospitals and the panel on p53 patterns was moderate to substantial (kappa 0.61), with about a quarter of p53 readings reclassified. Agreement among 13 panel pathologists on 316 biopsies was substantial (kappa 0.65). Adding p53 changed diagnoses: an aberrant pattern moved 9.2% of non-dysplastic biopsies to indefinite and 4.8% to low-grade dysplasia, and among biopsies indefinite for dysplasia, p53 moved 17.0% to non-dysplastic and 37.1% to low-grade dysplasia.

The indefinite-for-dysplasia category is where Barrett's surveillance decisions stall. A stain that resolves over half of those cases, with reproducible reading against defined patterns, is worth adopting where the expertise and quality control exist. The study does not show that the reclassified diagnoses predict progression better, which is the outcome that matters most.

• Consider p53 immunohistochemistry for Barrett's biopsies that are indefinite for dysplasia or show possible low-grade dysplasia.
• Report p53 using defined patterns: wild-type, overexpression, null, double clone or equivocal.
• Remember that null (complete loss) staining is aberrant, not negative.
• Seek expert or second review for low-grade dysplasia, which carries surveillance and treatment consequences.
• Train and audit against standardised p53 criteria; referring laboratories disagreed with the panel on about a quarter of readings.

Why it matters

Indefinite-for-dysplasia diagnoses leave patients in limbo; a reproducible stain moved more than half of them to a definite category.

Don't overread it

Reclassification is not the same as better prediction; the study did not follow patients for progression.

The statistics, in plain English

Kappa measures agreement beyond chance: 0.61 to 0.80 is conventionally called substantial. A kappa of 0.65 among 13 experts is good for a subjective histological read, but still means real disagreement in some cases.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.