PURPOSE: Assess whether cross-linking the carrier donor cornea of the Boston Keratoprosthesis (BKPro) improves retention of the device in participants at high risk of keratolysis. DESIGN: Prospective, double-masked randomized clinical trial. METHODS: In this multicenter study, sixty-eight adult participants who were scheduled for BKPro implantation were enrolled. Masked participants were randomized to receive either a cross-linked (CXL) or non-cross-linked (non-CXL) donor corneal carrier. Kaplan-Meier event-free survival was determined by the product-limit method and compared by the log-rank test to examine if survival curves were different between the CXL and non-CXL groups. The primary outcome of the study was time from surgery to BKPro removal. Secondary endpoint was twelve-month retention rate. RESULTS: Sixty-eight participants were enrolled and randomized 1:1 to each group. Average age at the time of surgery was 62 [24-89] years and 42 (62%) participants were male. Overall BKPro retention rate was 70% during a mean follow-up time of 93 (6 - 201) weeks. Twenty BKPros were removed, ten in the CXL group and ten in the non-CXL group, with 18 requiring removal because of sterile keratolysis. There was no difference in the time to removal between the groups during the study (P = 0.910). At twelve months, there was no significant difference in the retention rate in the CXL group (94%) versus the non-CXL group (82%, P = 0.150). CONCLUSION: In this prospective study, cross-linking of the carrier cornea prior to BKPro implantation did not reduce the incidence of sterile keratolysis or increase device retention among participants at high risk for retention failure.