PSA BASED EARLY DETECTION OF PROSTATE CANCER RECURRENCE

基于 PSA 的前列腺癌复发早期检测

• 批准号: 7303694
• 负责人: Jeremy M.G. Taylor
• 金额: $ 32.97万
• 依托单位: UNIVERSITY OF MICHIGAN AT ANN ARBOR
• 依托单位国家: 美国
• 财政年份: 2005
• 资助国家: 美国
• 起止时间: 2005-09-13 至 2008-12-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7303694
• 关键词: Adverse effects; Biological Assay; Cancer Center; Cancer Patient; Characteristics; Clinic; Clinical; Clinical Markers; Data; Data Set; Date Available; Early Diagnosis; Elevation; External Beam Radiation Therapy; General Hospitals; Heterogeneity; Internet; Intervention; Joints; Lead; Malignant Neoplasms; Malignant neoplasm of prostate; Massachusetts; Measurement; Measures; Methods; Michigan; Modeling; Monitor; Patients; Phase; Probability; Prostate-Specific Antigen; Radiation Therapy Oncology Group; Radiation therapy; Rate; Recording of previous events; Recurrence; Reproducibility; Scheme; Serum; Staging; Statistical Models; Time; Training; Uncertainty; Universities; University Hospitals; Work; base; cancer recurrence; follow-up; hormone therapy; programs; time use; tool; tumor; web based interface; web interface

The aim of this application is to develop a tool for feasible evidence-based clinical monitoring of prostate cancer patients treated by external beam radiation. Methods require the monitoring of serum prostate- specific antigen (PSA). Joint statistical models are fit to retrospective PSA and clinical recurrence data from 1945 patients treated at Massachusetts General Hospital and the University of Michigan Cancer Center (UMCC). Predictions of time to recurrence for new patients based on their current serial PSA history are obtained by integrating over fits from the joint model. The methods are validated using additional follow-up data from UMCC patients and from additional RTOG patients. Alternative prediction models are developed for patients treated with concurrent hormonal therapy using data from 541 UMCC patients and validated on data from RTOG. A web-based interface is constructed for access by clinicians and patients. The user enters characteristics of the patient along with all available dates and PSA values from start of treatment. The program returns an estimated probability of recurrence along with a measure of uncertainty. In addition to the early detection of prostate cancer recurrence, the long term objective of this work is to facilitate evidence-based monitoring of cancer patients through training of clinicians to interact with optimized programs on the world wide web.

此应用程序的目的是开发一种可行的循证临床监测前列腺的工具。 癌症患者接受外照射治疗。方法需要监测血清前列腺癌- 特异性抗原(PSA)。联合统计模型适用于回顾PSA和临床复发数据 1945名患者在马萨诸塞州总医院和密歇根大学癌症中心接受治疗 (UMCC)。根据新患者目前的连续PSA病史预测复发时间为 通过对关节模型进行过拟合积分而获得。这些方法通过额外的后续行动得到了验证 来自UMCC患者和其他RTOG患者的数据。开发了可供选择的预测模型 对于同时接受激素治疗的患者，使用来自541名UMCC患者的数据并验证 来自RTOG的数据。构建了一个基于Web的界面，以供临床医生和患者访问。用户 输入患者的特征以及自治疗开始以来的所有可用日期和PSA值。 该程序返回估计的复发概率以及不确定性的度量。此外 为了早期发现前列腺癌的复发，这项工作的长期目标是促进 循证监测癌症患者通过培训临床医生与优化互动 万维网上的节目。