Objective To observe the different clinical response patterns of uveal melanoma (UM) patients after external scleral plaque radiotherapy (PRT), and to investigate the risk factors of secondary enucleation after treatment failure. MethodsA single-centre retrospective study. Demographic baseline characteristics and clinical data were collected from 465 UM patients treated with 125I external scleral PRT at Beijing Tongren Hospital from March 2011 to September 2017. Among them, 217 were male and 248 were female, tumor all occurred monocularly. The mean age of subjects was 46.7±12.1 years. Reasons for secondary enucleation included local tumor treatment failure, glaucoma, scleral necrosis and patient request. Tumor grading was based on the grading standards established by the American Joint Committee on Cancer (AJCC). The pattern of tumor response after PRT was classified as degenerated type, growth type, stable type or other types according to literature criteria. The median follow-up time after PRT was 59 months to observe tumor changes. Complete follow-up records of 3 or more color doppler ultrasound imaging (CDI) was available in 245 cases. A t-test was performed to compare the patient's age, intraocular pressure, best corrected visual acuity, tumor thickness and maximum basal diameter before treatment; a chi-square test was performed to compare the patient's gender, AJCC T classification of the tumor, whether the ciliary body was involved, presence of subretinal fluid, optic disc invasion and vitreous hemorrhage, tumor shape and location. Kaplan-Meier survival analysis was used to estimate the cumulative probability of secondary enucleation after extra-scleral PRT. Univariate and multivariate Cox proportional hazards regression analyses were used to evaluate the relationship between tumor characteristics and secondary enucleation after extra-scleral PRT. ResultsAmong 465 patients, eecondary enucleation was performed on 78 (16.8%, 78/465) patients during the follow-up period. The 1, 3 and 5 year secondary enucleation rates were 5.4%, 9.3% and 17.1%, respectively. Eye preservation was successful in 387 cases (83.2%, 387/465). Patients treated by secondary enucleation had a larger maximum basal diameter of tumor, a higher proportion of irregular and diffuse morphology, a cumulative macular involved and a higher AJCC T classification, the difference was statistically significant (P＜0.05). There were 115, 76, 27, and 27 cases of degenerated type, stable type, growth type, and other type, respectively. The tumor thickness of the growth type and other types was significantly smaller than that of the degenerated type and the stable type, and the difference was statistically significant (P＜0.05). Univariate Cox analysis showed that the maximum basal diameter of the tumor (HR=1.19), tumor thickness (HR=1.08), AJCC T classification (HR=1.90), growth type response pattern (relative to degenerated type response pattern) (HR=4.20) was associated with failure of eye preservation (P＜0.05). In the multivariate Cox analysis, the largest tumor basal diameter (HR=1.24) and the growth type response pattern (relative to the degenerated type response pattern) (HR=4.59) were still associated with failure of eye preservation (P＜0.05). ConclusionsThe tumor thickness of UM patients with growing and other response patterns after PRT is smaller before treatment; the maximum basal diameter of the tumor and the growing response pattern are independent risk factors for secondary enucleation.