Although generally well tolerated, oral aminobisphosphonates have been associated with sporadic cases of severe esophageal injury attributed to pill contact. A novel protected formulation of oral alendronate uses an inert cylindrical shell to prevent mucosal contact with intact tablets. An active and placebo-controlled endoscopy study was undertaken to assess mucosal injury associated with this protected formulation. Healthy volunteers with normal baseline endoscopy were randomly assigned to receive protected alendronate 70 mg/day, standard alendronate 70 mg/day, or placebo for 14 days. Endoscopy was repeated on days 8 and 15. Of 78 subjects, 30 received protected alendronate, 28 received standard alendronate, and 20 received placebo. Mean gastric injury scores did not differ significantly among treatment groups. However, subjects on standard alendronate were more likely than those on protected alendronate to develop severe gastric injury, defined as Lanza score 3 or 4 (67.9% versus 33.3%, P=.009), and more likely to develop a gastric ulcer (21.4% versus 3.3%, P=.015). No differences in symptoms or adverse events were observed. In conclusion, a novel protected formulation of oral alendronate is less likely than standard alendronate to induce severe mucosal injury to the upper gastrointestinal tract.