在患有脉络膜黑色素瘤的患者中，局部复发（LR）造成的疾病特异性死亡率增加，但其潜在机制尚不清楚。本研究旨在确定诊断时染色体3和8q的拷贝数变异是否增加LR的发生率，并且在患有LR的患者中，疾病特异性死亡率是否取决于原发肿瘤的染色体状态。本研究是一个回顾性队列研究。受试者共239名，包括2009年1月至2019年12月在一个单一的国家转诊中心接受Ru-106近距离放射治疗的原发性脉络膜黑色素瘤患者。使用Cox回归模型和Kaplan-Meier分析评估染色体3和8q状态对LR和LR事件后疾病特异性死亡率的影响。使用多状态模型说明患者随时间的概率：'存活且无疾病'，'存活且LR'，'死于脉络膜黑色素瘤转移'，或'死于其他原因'，并根据染色体3和8q状态分组。 结果显示，42名患者（16％）发生了LR。总的LR发生率并不受染色体3和8q状态的影响（p = 0.87）。与染色体3和8q拷贝数正常的患者相比，携带染色体3和/或8q异常的患者LR发生时间更早，原发诊断至LR的中位数时间分别为1.6年（IQR：1.0-2.0）和3.2年（IQR：2.1-5.0）。Cox回归分析发现LR是疾病特异性死亡率的独立风险因素（所有患者的HR为2.7（95％CI 1.5-5.0）），但多状态模型表明，即使在发生LR后，拷贝数正常的染色体3和8q患者的疾病特异性死亡风险非常低。原发脉络膜黑色素瘤的染色体3或8q拷贝数变异并没有增加总体LR发生率。然而，LR的发生增加了携带染色体3和/或8q异常患者的疾病特异性死亡风险。即使在LR后，染色体3和8q拷贝数正常的患者的疾病特异性死亡率仍然很低。版权所有©2023年，由Elsevier Inc.发布。

Increased disease-specific mortality has been observed among patients with local recurrence (LR) from uveal melanoma, but the underlying mechanism is unknown. The purpose of this study was to determine if copy number alterations of chromosomes 3 and 8q, at the time of diagnosis, increase the incidence of LR and if disease-specific mortality among patients with LR depends on the chromosome status of the primary tumor.Retrospective cohort study SUBJECTS: The study included 239 consecutive patients with primary uveal melanoma (choroidal or ciliary body) treated with Ru-106 brachytherapy, from January 2009 to December 2019 at a single national referral center.Cox-regression modelling and Kaplan-Meier analyses were used to assess the effect of chromosomes 3 and 8q status on the incidence of LR and disease-specific mortality following the event of LR. Multistate models were used to illustrate the probabilities over time of patients being: 'alive and disease-free', 'alive with LR', 'dead from uveal melanoma metastases', or 'dead from other causes' split on chromosomes 3 and 8q status.LR incidence and disease-specific mortality RESULTS: LR was observed in 42 patients (16%). Overall incidence of LR was not affected by chromosome 3 and 8q status (p=0.87), albeit LR occurred earlier in patients with aberrations of chromosomes 3 and/or 8q compared to patients with normal copy number of chromosomes 3 and 8q, median time from primary diagnosis to LR was 1.6 years (IQR: 1.0-2.0) and 3.2 years (IQR 2.1-5.0), respectively. Cox regression found LR to be an independent risk factor for disease-specific mortality (HR 2.7 (95%CI 1.5-5.0) among all patients, but multistate models demonstrated very low risk of disease-specific death among patients with normal chromosomes 3 and 8q status, even following a LR.Copy number alterations of chromosomes 3 or 8q in the primary uveal melanoma did not increase the overall incidence of LR. However, the development of a LR enhanced the risk of disease-specific mortality among patients with copy number alterations of chromosomes 3 and/or 8q. Even after a LR, disease-specific mortality remained low among patients with normal copy numbers of chromosomes 3 and 8q.Copyright © 2023. Published by Elsevier Inc.