Topically applied methylene blue dye chromoendoscopy is effective in improving detection of colorectal neoplasia. When combined with a pH- and time-dependent multimatrix structure, a per-oral methylene blue formulation (MB-MMX) can be delivered directly to the colorectal mucosa.

We performed a phase 3 study of 1205 patients scheduled for colorectal cancer screening or surveillance colonoscopies (50–75 years old) at 20 sites in Europe and the United States, from December 2013 through October 2016. Patients were randomly assigned to groups given 200 mg MB-MMX, placebo, or 100 mg MB-MMX (ratio of 2:2:1). The 100-mg MB-MMX group was included for masking purposes. MB-MMX and placebo tablets were administered with a 4-L polyethylene glycol–based bowel preparation. The patients then underwent colonoscopy by an experienced endoscopist with centralized double-reading. The primary endpoint was the proportion of patients with 1 adenoma or carcinoma (adenoma detection rate [ADR]). We calculated odds ratios (ORs) and 95% confidence intervals (CIs) for differences in detection between the 200-mg MB-MMX and placebo groups. False-positive (resection rate for non-neoplastic polyps) and adverse events were assessed as secondary endpoints.

The ADR was higher for the MB-MMX group (273 of 485 patients, 56.29%) than the placebo group (229 of 479 patients, 47.81%) (OR 1.46; 95% CI 1.09–1.96). The proportion of patients with nonpolypoid lesions was higher in the MB-MMX group (213 of 485 patients, 43.92%) than the placebo group (168 of 479 patients, 35.07%) (OR 1.66; 95% CI 1.21–2.26). The proportion of patients with adenomas ≤5 mm was higher in the MB-MMX group (180 of 485 patients, 37.11%) than the placebo group (148 of 479 patients, 30.90%) (OR 1.36; 95% CI 1.01–1.83), but there was no difference between groups in detection of polypoid or larger lesions. The false-positive rate did not differ significantly between groups (83 [23.31%] of 356 patients with non-neoplastic lesions in the MB-MMX vs 97 [29.75%] of 326 patients with non-neoplastic lesions in the placebo group). Overall, 0.7% of patients had severe adverse events but there was no significant difference between groups.

In a phase 3 trial of patients undergoing screening or surveillance colonoscopies, we found MB-MMX led to an absolute 8.5% increase in ADR, compared with placebo, without increasing the removal of non-neoplastic lesions. Clinicaltrials.gov no: NCT01694966