文献速递中枢神经肿瘤放疗泌尿生殖与妇科肿瘤放疗放疗物理与技术

医生主导的多准则优化计划:前列腺癌与脑肿瘤的试点研究

Multicriteria plan optimization in the hands of physicians: a pilot study in prostate cancer and brain tumors.
PubMed 文献 · Radiat Oncol · 2017-11-06

该试点研究回顾性对12例脑肿瘤和10例前列腺切除术后患者,由医生使用基于模板的多准则优化(MCO)导航生成IMRT计划,并与物理师临床计划比较。结果显示计划质量相似,医生计划在前列腺癌高剂量直肠和膀胱区域保护更优,脑肿瘤靶区和脑干剂量更高;盲法比较中临床计划更常被偏好但无统计学差异,脑病例医生参与时间减少30%,物理师时间均减少。

要点

  • 医生使用MCO导航可生成与物理师临床计划质量相似的IMRT计划。
  • 前列腺癌医生计划在高剂量直肠和膀胱区域保护显著更优(p(D1)<0.05)。
  • 脑肿瘤医生计划靶区和脑干剂量更高(p(D1)<0.05),脑病例医生参与时间减少30%。

核心结论

  • 医生主导的MCO计划工作流在脑肿瘤和前列腺癌中可行,计划质量与临床计划相当,并可能提高科室效率。

研究设计与样本

  • 回顾性试点研究,对12例脑肿瘤和10例前列腺切除术后前列腺癌患者,由医生使用基于模板的MCO Pareto面导航生成计划,并与物理师临床计划比较剂量学差异、医生偏好和计划时间。

主要发现

  • 医生计划与临床计划总体质量相似,但临床目标优先级不同。
  • 医生前列腺计划在高剂量直肠和膀胱区域保护显著更优(p(D1)<0.05)。
  • 医生脑肿瘤计划靶区和脑干剂量更高(p(D1)<0.05)。
  • 盲法计划比较中医生更常偏好临床计划(脑6:3/12,前列腺2:6/10),无统计学显著性。
  • 脑病例医生平均参与时间减少30%,所有病例物理师时间均减少。

关键数值与推荐

  • 前列腺癌医生计划高剂量直肠和膀胱区域保护显著更优,p(D1)<0.05,D1为对应结构1%体积接受的剂量。
  • 脑肿瘤医生计划靶区和脑干剂量更高,p(D1)<0.05。
  • 盲法偏好:脑肿瘤临床计划6:3(共12例),前列腺癌临床计划2:6(共10例),无统计学显著性。
  • 脑病例医生平均参与时间减少30%。

临床怎么用

  • 可利用站点特异性模板和MCO导航让医生更深入参与计划优化,标准化计划流程。
  • 在脑肿瘤和前列腺癌中,医生导航计划可作为可行的工作流,但需注意脑肿瘤靶区和脑干剂量可能更高。
  • 该工作流可减少物理师计划时间,脑病例还可减少医生参与时间,可能提升科室效率。

局限性

  • 回顾性试点研究,样本量有限(12例脑肿瘤和10例前列腺癌)。
  • 盲法计划比较中偏好差异无统计学显著性。
  • 主观获益如对计划优化和剂量学权衡的理解为观察性结果。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: The purpose of this study was to demonstrate the feasibility of physician driven planning in intensity modulated radiotherapy (IMRT) with a multicriteria optimization (MCO) treatment planning system and template based plan optimization. Exploiting the full planning potential of MCO navigation, this alternative planning approach intends to improve planning efficiency and individual plan quality. METHODS: Planning was retrospectively performed on 12 brain tumor and 10 post-prostatectomy prostate patients previously treated with MCO-IMRT. For each patient, physicians were provided with a template-based generated Pareto surface of optimal plans to navigate, using the beam angles from the original clinical plans. We compared physician generated plans to clinically delivered plans (created by dosimetrists) in terms of dosimetric differences, physician preferences and planning times. RESULTS: Plan qualities were similar, however physician generated and clinical plans differed in the prioritization of clinical goals. Physician derived prostate plans showed significantly better sparing of the high dose rectum and bladder regions (p(D1) < 0.05; D1: dose received by 1% of the corresponding structure). Physicians' brain tumor plans indicated higher doses for targets and brainstem (p(D1) < 0.05). Within blinded plan comparisons physicians preferred the clinical plans more often (brain: 6:3 out of 12, prostate: 2:6 out of 10) (not statistically significant). While times of physician involvement were comparable for prostate planning, the new workflow reduced the average involved time for brain cases by 30%. Planner times were reduced for all cases. Subjective benefits, such as a better understanding of planning situations, were observed by clinicians through the insight into plan optimization and experiencing dosimetric trade-offs. CONCLUSIONS: We introduce physician driven planning with MCO for brain and prostate tumors as a feasible planning workflow. The proposed approach standardizes the planning process by utilizing site specific templates and integrates physicians more tightly into treatment planning. Physicians' navigated plan qualities were comparable to the clinical plans. Given the reduction of planning time of the planner and the equal or lower planning time of physicians, this approach has the potential to improve departmental efficiencies.

来源:Radiat Oncol,查看原文。版权归原作者所有。

引用

  1. Müller BS, Shih HA, Efstathiou JA, et al. Multicriteria plan optimization in the hands of physicians: a pilot study in prostate cancer and brain tumors. Radiat Oncol. 2017. PMID: 29110689. doi:10.1186/s13014-017-0903-z

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。