We previously demonstrated the efficacy of BNCT mediated by boronophenylalanine (BPA) to treat tumors in a hamster cheek pouch model of oral cancer with no normal tissue radiotoxicity and moderate, albeit reversible, mucositis in precancerous tissue around treated tumors. It is known that boron targeting of the largest possible proportion of tumor cells contributes to the success of BNCT and that tumor blood vessel normalization improves drug delivery to the tumor. Within this context, the aim of the present study was to evaluate the effect of blood vessel normalization on the therapeutic efficacy and potential radiotoxicity of BNCT in the hamster cheek pouch model of oral cancer. Blood vessel normalization was induced by two doses of thalidomide in tumor-bearing hamsters on 2 consecutive days. All studies in thalidomide-treated animals were performed 48 h after the first dose of thalidomide, previously established as the window of normalization. Biodistribution studies were performed with BPA at a dose of 15.5 mg ¹⁰B/kg in thalidomide-treated (Th ) and untreated (Th–) tumor-bearing hamsters. The effect of blood vessel normalization prior to BPA administration on the efficacy of BNCT was assessed in in vivo BNCT studies at the RA-3 Nuclear Reactor in tumor-bearing hamsters. Group I was treated with BPA-BNCT after treatment with thalidomide (Th BPA-BNCT). Group II was treated with BPA-BNCT alone (Th– BPA-BNCT). Group III was treated with the beam only after treatment with thalidomide (Th BO), and Group IV was treated with the beam only (Th– BO). Groups I and II were given the same dose of BPA (15.5 mg ¹⁰B/kg), and all groups (I–IV) were exposed to the same neutron fluence. Two additional groups were treated with the beam only at a higher dose to exacerbate mucositis in precancerous tissue and to explore the potential direct protective effect of thalidomide on radiation-induced mucositis in a scenario of more severe toxicity, i.e. Group V (Th hdBO) and Group VI (Th– hdBO). The animals were followed for 28 days. Biodistribution studies revealed no statistically significant differences in gross boron content between Th and Th– animals. Overall tumor control (complete response partial response) at 28 days post-treatment was significantly higher for Group I (Th BPA-BNCT) than for Group II (Th– BPA-BNCT): 84 ± 3% compared to 67 ± 5%. Pretreatment with thalidomide did not induce statistically significant changes in overall tumor control induced by the beam only, i.e. 15 ± 5% in Group III (Th BO) and 18 ± 5% in Group IV (Th– BO), or in overall tumor control induced by the high-dose beam only, i.e. 60 ± 7% in Group V (Th hdBO) and 47 ± 10% in Group VI (Th– hdBO). BPA-BNCT alone (Group II) induced mucositis in precancerous tissue that reached Grades 3–4 in 80% of the animals, whereas pretreatment with thalidomide (Group I) prevented mucositis Grades 3 and 4 completely. Beam-only Group III (Th BO) exhibited only Grade 1 mucositis in precancerous tissue, whereas 17% of the animals in beam-only Group IV (Th– BO) reached Grade 2 mucositis. High-dose beam-only group V (Th hdBO) exhibited only Grade 2 mucositis, whereas high-dose beam-only group VI (Th– hdBO) reached Grade 3 mucositis in 83% of the animals. In all cases mucositis in precancerous tissue was reversible. No normal tissue radiotoxicity was observed with any of the protocols. Pretreatment with thalidomide enhanced the therapeutic efficacy of BNCT and reduced precancerous tissue toxicity.