Abe James, Lobo Jennifer M, Trifiletti Daniel M, Showalter Timothy N
Department of Radiation Oncology, University of Virginia School of Medicine, 1240 Lee Street, Box 800383, Charlottesville, VA, 22908, USA.
Department of Public Health Sciences, University of Virginia School of Medicine, Charlottesville, VA, USA.
BMC Med Inform Decis Mak. 2017 Aug 24;17(1):128. doi: 10.1186/s12911-017-0526-1.
Despite the emergence of genomics-based risk prediction tools in oncology, there is not yet an established framework for communication of test results to cancer patients to support shared decision-making. We report findings from a stakeholder engagement program that aimed to develop a framework for using Markov models with individualized model inputs, including genomics-based estimates of cancer recurrence probability, to generate personalized decision aids for prostate cancer patients faced with radiation therapy treatment decisions after prostatectomy.
We engaged a total of 22 stakeholders, including: prostate cancer patients, urological surgeons, radiation oncologists, genomic testing industry representatives, and biomedical informatics faculty. Slides were at each meeting to provide background information regarding the analytical framework. Participants were invited to provide feedback during the meeting, including revising the overall project aims. Stakeholder meeting content was reviewed and summarized by stakeholder group and by theme.
The majority of stakeholder suggestions focused on aspects of decision aid design and formatting. Stakeholders were enthusiastic about the potential value of using decision analysis modeling with personalized model inputs for cancer recurrence risk, as well as competing risks from age and comorbidities, to generate a patient-centered tool to assist decision-making. Stakeholders did not view privacy considerations as a major barrier to the proposed decision aid program. A common theme was that decision aids should be portable across multiple platforms (electronic and paper), should allow for interaction by the user to adjust model inputs iteratively, and available to patients both before and during consult appointments. Emphasis was placed on the challenge of explaining the model's composite result of quality-adjusted life years.
A range of stakeholders provided valuable insights regarding the design of a personalized decision aid program, based upon Markov modeling with individualized model inputs, to provide a patient-centered framework to support for genomic-based treatment decisions for cancer patients. The guidance provided by our stakeholders may be broadly applicable to the communication of genomic test results to patients in a patient-centered fashion that supports effective shared decision-making that represents a spectrum of personal factors such as age, medical comorbidities, and individual priorities and values.
尽管肿瘤学领域出现了基于基因组学的风险预测工具，但尚未建立一个将检测结果传达给癌症患者以支持共同决策的既定框架。我们报告了一项利益相关者参与计划的结果，该计划旨在开发一个框架，使用马尔可夫模型及个性化模型输入（包括基于基因组学的癌症复发概率估计），为前列腺切除术后面临放射治疗决策的前列腺癌患者生成个性化决策辅助工具。
我们共邀请了22名利益相关者，包括：前列腺癌患者、泌尿外科医生、放射肿瘤学家、基因组检测行业代表和生物医学信息学教员。每次会议都展示幻灯片，提供有关分析框架的背景信息。邀请参与者在会议期间提供反馈，包括修改总体项目目标。利益相关者会议内容按利益相关者群体和主题进行了审查和总结。
大多数利益相关者的建议集中在决策辅助工具的设计和格式方面。利益相关者对使用具有个性化模型输入的决策分析模型来评估癌症复发风险以及年龄和合并症带来的竞争风险的潜在价值充满热情，认为这可以生成一个以患者为中心的工具来辅助决策。利益相关者不认为隐私问题是拟议的决策辅助计划的主要障碍。一个共同的主题是，决策辅助工具应能在多个平台（电子和纸质）上使用，应允许用户进行交互以迭代调整模型输入，并在会诊预约前和会诊期间提供给患者。重点强调了解释模型的质量调整生命年综合结果所面临的挑战。
一系列利益相关者就基于具有个性化模型输入的马尔可夫建模的个性化决策辅助计划的设计提供了宝贵见解，以提供一个以患者为中心的框架，支持癌症患者基于基因组学的治疗决策。我们的利益相关者提供的指导可能广泛适用于以患者为中心的方式向患者传达基因组检测结果，以支持有效的共同决策，该决策代表了一系列个人因素，如年龄、医疗合并症以及个人优先事项和价值观。
