We previously demonstrated the efficacy of boron neutron capture therapy (BNCT) mediated by boronophenylalanine (BPA), GB-10 (Na(2)(10)B(10)H(10)) and (GB-10+BPA) to control tumors, with no normal tissue radiotoxicity, in the hamster cheek pouch oral cancer model. Herein we developed a novel experimental model of field-cancerization and precancerous lesions (globally termed herein precancerous tissue) in the hamster cheek pouch to explore the long-term potential inhibitory effect of the same BNCT protocols on the development of second primary tumors from precancerous tissue. Clinically, second primary tumor recurrences occur in field-cancerized tissue, causing therapeutic failure. We performed boron biodistribution studies followed by in vivo BNCT studies, with 8 months follow-up. All 3 BNCT protocols induced a statistically significant reduction in tumor development from precancerous tissue, reaching a maximum inhibition of 77-100%. The inhibitory effect of BPA-BNCT and (GB-10+BPA)-BNCT persisted at 51% at the end of follow-up (8 months), whereas for GB-10-BNCT it faded after 2 months. Likewise, beam-only elicited a significant but transient reduction in tumor development. No normal tissue radiotoxicity was observed. At 8 months post-treatment with BPA-BNCT or (GB-10+BPA)-BNCT, the precancerous pouches that did not develop tumors had regained the macroscopic and histological appearance of normal (non-cancerized) pouches. A potential new clinical application of BNCT would lie in its capacity to inhibit local regional recurrences.