Description

The investigators will perform a prospective, multicenter, randomized controlled trial of 2 different gastric biopsy protocols. The hypothesis is that a gastric mapping protocol not including the incisura biopsy ("I-": 2 biopsies from the antrum, 2 from the corpus) is non-inferior to the conventional 5-biopsy approach ("I+": 2 biopsies from the antrum, 2 from the corpus, 1 from the incisura). The primary endpoint of this study is to assess the sensitivity of the "I-" protocol in diagnosing high-risk precancerous chronic gastritis (OLGA/OLGIM stages III or IV) compared to the assumed gold standard of histology based on the "I+" protocol.

In designing the study, the investigators opted for a non-inferiority paired design with high-risk chronic gastritis as a primary binary outcome and a non-inferiority margin (Δ) of 5 % (two-sided 95% CI). The non-inferiority margin was established according to clinical judgement, taking into consideration the significance of a missed diagnosis of high-risk preneoplastic condition and previous estimates of the added value offered by the added incisura biopsy. With a power of 80% and a 2-sided α-level set at 0.05, it is estimated that 370 subjects are needed to show non-inferiority. Finally, 400 subjects will be enrolled to compensate for eventual exclusions or protocol deviations.

Cohen's kappa will be used to rate intra and inter-observer concordance for pathologists.

An intermediary analysis for futility is scheduled after 100 patients (25%) are enrolled.