指南与共识泌尿生殖技术与安全胸部PETCTMR

PSMA-PET不确定骨病灶与淋巴结多学科解读算法

Evaluating indeterminate bone lesions and lymph nodes on PSMA-PET: a multidisciplinary consensus algorithm and 1-year implementation results.
指南 / 共识 · Eur Radiol · 2026-01-06

针对前列腺癌PSMA-PET中不确定骨病灶和淋巴结,本研究制定并前瞻性实施多学科共识算法,整合PSMA-PET特征、常规影像、临床病理信息。2024—2025年共81例患者经MDT讨论,算法判定为良性48.1%、不确定4.9%、转移40.7%、混合6.2%。与复合参考标准比较,算法在可评估患者中94.1%判断正确,并较PSMA-RADS及PROMISE V2 PSMA表达评分明显减少不确定解读。

要点

  • 整合临床病理与常规影像可明显降低PSMA-PET不确定骨病灶和淋巴结比例。
  • MDT共同阅片和算法化分级有助于直接形成治疗决策。
  • PSMA-RADS与PROMISE V2在MDT场景中仍可产生较多不确定判读。

核心结论

  • 本多学科共识算法整合临床病理信息和常规影像,在MDT场景下使PSMA-PET不确定骨病灶和淋巴结的不确定判读频率降至4.9%。
  • 在具有充分参考标准的68例患者中,算法判断正确率为94.1%(64/68)。
  • 算法与PSMA-RADS在不一致率为54.3%(44/81),与PROMISE V2 PSMA表达评分不一致率为71.6%(58/81)。

研究设计与样本

  • 回顾性单中心研究,评估一个前瞻性实施的多学科共识算法。
  • 纳入2024—2025年连续81例前列腺癌患者,均为MDT讨论PSMA-PET不确定骨病灶或淋巴结者。
  • 患者中位年龄68岁,四分位距64—75岁。
  • 以复合参考标准作为结果对照,并与PSMA-RADS和PROMISE V2 PSMA表达评分进行比较。

关键影像表现/征象

  • 算法综合考虑病灶位置、常规影像特征、PSMA-PET特征及临床病理信息。

诊断标准与关键数值

  • 算法结果分布:良性48.1%(39/81),不确定4.9%(4/81),转移40.7%(33/81),混合(良性及转移病灶并存)6.2%(5/81)。
  • 算法正确率为94.1%(64/68例具备充分参考标准的患者)。
  • PSMA-RADS不确定比例为53.1%(43/81),PROMISE V2 PSMA表达评分不确定比例为64.2%(52/81)。
  • 算法不确定比例为4.9%(4/81),低于PSMA-RADS和PSMA表达评分。

临床应用与阅片要点

  • 对PSMA-PET不确定骨病灶或淋巴结,应在MDT中结合PSMA-PET征象、常规影像和临床病理信息分级。
  • 使用结构化多学科算法可减少报告中的不确定结论,便于后续治疗决策。
  • 有常规影像形态学支持或临床病理佐证时,应主动归类为良恶性倾向,而非简单报告不确定。

局限性

  • 回顾性单中心研究设计。
  • 部分患者缺乏充分参考标准。
  • 原文未提供观察者间一致性或算法组分权重验证。

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)。

英文原文(摘要)· 点击展开

OBJECTIVE: Indeterminate lesions on prostate-specific membrane antigen (PSMA)-PET are challenging to address. We aimed to develop, implement, and evaluate a multidisciplinary consensus algorithm that integrates existing interpretation systems with multimodality imaging and clinicopathological information for interpreting indeterminate bone and lymph node lesions on PSMA-PET.

MATERIALS AND METHODS: This was a retrospective single-center study on a prospectively implemented algorithm. We included all consecutive prostate cancer patients whose PSMA-PET findings for indeterminate bone lesions or lymph nodes were discussed at a multidisciplinary tumor board (MDT) in 2024-2025. An algorithm determining the level of suspicion for metastasis was developed in a multidisciplinary fashion, incorporating lesion location, conventional imaging features, PSMA-PET characteristics, and clinicopathological information. The application of the algorithm and outcomes were documented, compared against a composite reference standard. Comparisons were made with PSMA-RADS and PROMISE V2 PSMA-expression scores.

RESULTS: 81 patients (median age 68, interquartile range 64-75) were included. Algorithm results were benign (48.1% [39/81]), equivocal (4.9% [4/81]), metastasis (40.7% [33/81]), and mixed (benign and metastatic lesions, 6.2% [5/81]). The algorithm was correct in 94.1% (64 of 68 patients with a sufficient reference standard). The algorithm was discordant with PSMA-RADS in 54.3% (44/81) and with PROMISE V2 PSMA-expression score in 71.6% (58/81). The frequency of equivocal lesions was lower using the algorithm (4.9% [4/81]) compared with PSMA-RADS (53.1% [43/81]) and PSMA-expression score (64.2% [52/81]).

CONCLUSION: A multidisciplinary consensus algorithm for interpreting indeterminate bone lesions and lymph nodes on PSMA-PET was developed and implemented. Integrating clinicopathological information and multimodality imaging in an MDT setting reduced equivocal interpretations.

KEY POINTS: Question While prostate-specific membrane antigen (PSMA)-PET has become essential in the management of prostate cancer, indeterminate bone lesions and lymph nodes remain challenging to address. Findings A multidisciplinary algorithm for interpreting indeterminate bone lesions and lymph nodes on PSMA-PET, incorporating clinicopathological information and multimodality imaging, reduced the frequency of equivocal interpretations. Clinical relevance An algorithm for interpreting indeterminate bone lesions and lymph nodes on PSMA-PET, incorporating clinicopathological information and multimodality imaging in a multidisciplinary tumor board setting, decreases the frequency of equivocal interpretations and can potentially help management decisions.

来源:Eur Radiol,查看原文。版权归原作者所有。

引用

  1. Woo, Tong, Becker, et al. Evaluating indeterminate bone lesions and lymph nodes on PSMA-PET: a multidisciplinary consensus algorithm and 1-year implementation results. Eur Radiol. 2026. PMID: 41493546. doi:10.1007/s00330-025-12301-w

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。