本研究的目标是开发一个风险预测模型，用于估计被诊断为第四期乳腺癌的年轻女性的总体生存率。临床信息来源于2010年至2015年间的监测、流行病学和终结果（SEER）数据库。为了确定依赖的风险因素，我们分别使用了单变量和多变量的Cox比例风险回归模型进行分析。然后，我们基于确定的风险因素创建了一个新的示意图，以预测患者的1年、3年和5年总体生存率概率。符合资格要求的676名患者以随机方式划分为训练组（n = 475）和验证组（n = 201），比例为7:3。组织学、乳腺亚型、T分期、脑转移、骨转移、肝转移和手术被确定为癌症的独立预后因素。为了预测1年、3年和5年总体生存（OS）的概率，所有这些独立因素都被纳入到示意图中。我们的示意图表现出良好的鉴别能力，训练集和验证集的C指数分别为0.737（95% CI: 0.708-0.766）和0.717（95% CI: 0.664-0.770）。标定曲线显示出两个队列的良好一致性。利用这个示意图，我们开发了一个风险分层模型，将患者分为低风险、中风险和高风险组。这个预测模型更精确地预测了第四期乳腺癌年轻女性的总体生存率，并且能够进行个体化的风险估计，有助于医生探索治疗策略的有效性。版权所有©2023年Elsevier Inc.。保留所有权利。

The goal of this study is to develop a risk prediction model for estimating overall survival (OS) in young females diagnosed with stage IV breast cancer.The clinical information was retrieved from the Surveillance, Epidemiology, and End Results (SEER) database between 2010 and 2015. To identify the dependent risk factors, we utilized the Cox proportional hazards regression model in both single and multivariate analyses. We then created a new nomogram to predict the 1-, 3-, and 5-year overall survival probability for these patients based on the identified risk factors.Six hundred seventy-six patients who met the eligibility requirements were stochastically partitioned into training (n = 475) and validation (n = 201) groups in a 7:3 ratio. Histology, breast subtype, T classification, brain metastasis, bone metastasis, liver metastasis, and surgery were identified as independent prognostic factors for cancer. To predict the 1-, 3-, and 5-year overall survival (OS) probabilities, all of these independent factors were incorporated into nomograms. Our nomogram demonstrated a favorable discriminatory power, as evidenced by a C-index of 0.737 (95% CI: 0.708-0.766) and 0.717 (95% CI: 0.664-0.770) for the training and validation cohorts, respectively. The calibration curves showed satisfactory consistency in both cohorts. Using this nomogram, we developed a risk stratification model that categorized patients into low-, intermediate-, and high-risk groups.The prediction model was more precisely to predict the OS of young females with stage IV breast cancer and could enable individualized risk estimation that could be conducive to physicians exploring therapeutic strategies for effectiveness.Copyright © 2023 Elsevier Inc. All rights reserved.