文献速递泌尿生殖腹部MR

单次激发T1FLASH序列前列腺T1 mapping的临床可行性研究

T1 Mapping of the Prostate Using Single-Shot T1FLASH: A Clinical Feasibility Study to Optimize Prostate Cancer Assessment.
PubMed 文献 · Invest Radiol · 2022-12-13

前瞻性纳入40例接受超声/MRI融合引导穿刺的患者,在3T扫描仪上使用单次激发T1FLASH技术行前列腺T1 mapping,中位采集时间2:10分钟且无明显伪影。T1弛豫时间在移行区和外周区均能区分前列腺癌与参考区及良性增生结节,鉴别前列腺癌与影像可疑但病理良性病变的AUC为0.69,与ADC的0.62无显著差异。

要点

  • T1FLASH前列腺T1 mapping可在2:10分钟内完成多层面采集,图像无明显伪影,临床可行。
  • 移行区前列腺癌T1弛豫时间显著低于参考区和良性增生结节(P=0.029和P<0.001)。
  • 外周区前列腺癌T1弛豫时间显著低于参考区(P<0.001)。

关键影像表现/征象

  • 移行区前列腺癌T1弛豫时间较参考区缩短(P=0.029),较良性增生结节缩短(P<0.001)。
  • 外周区前列腺癌T1弛豫时间较参考区缩短(P<0.001)。
  • 前列腺癌T1弛豫时间中位数1.40秒,影像可疑但病理良性病变中位数1.47秒,差异未达统计学显著(P=0.066)。

诊断标准与关键数值

  • 鉴别前列腺癌与参考区及良性前列腺病变:T1弛豫时间AUC=0.80,ADC值AUC=0.83,两者比较P=0.519。
  • 鉴别前列腺癌与影像可疑但病理良性病变:T1弛豫时间AUC=0.69,ADC值AUC=0.62,两者比较P=0.446。
  • 临床显著前列腺癌定义为ISUP分级≥2级,22例前列腺癌中ISUP 1/2/3/4级分别为3/15/3/1例。

临床应用与阅片要点

  • T1 mapping可作为多参数MRI的补充序列,短时间采集不增加明显伪影,便于临床整合。
  • 阅片时注意移行区前列腺癌T1弛豫时间可低于良性增生结节,但鉴别影像可疑良性病变时效能中等(AUC=0.69)。

局限性

  • 单中心前瞻性研究,最终仅纳入40例参与者,样本量有限。
  • 作者指出需更大样本验证,并评估T1FLASH与深度学习结合及与扩散加权成像的稳健性比较。

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

PURPOSE: The aim of this study was to assess the clinical feasibility of magnetic resonance imaging (MRI) T1 mapping using T1FLASH for assessment of prostate lesions.

METHODS: Participants with clinical suspicion for prostate cancer (PCa) were prospectively enrolled between October 2021 and April 2022 with multiparametric prostate MRI (mpMRI) acquired on a 3 T scanner. In addition, T1 mapping was accomplished using a single-shot T1FLASH technique with inversion recovery, radial undersampling, and iterative reconstruction. Regions of interest (ROIs) were manually placed on radiologically identified prostate lesions and representative reference regions of the transitional zone (TZ), benign prostate hyperplasia nodules, and peripheral zone (PZ). Mean T1 relaxation times and apparent diffusion coefficient (ADC) values (b = 50/b = 1400 s/mm 2 ) were measured for each ROI. Participants were included in the study if they underwent ultrasound/MRI fusion-guided prostate biopsy for radiologically or clinically suspected PCa. Histological evaluation of biopsy cores served as reference standard, with grading of PCa according to the International Society of Urological Pathology (ISUP). ISUP grades 2 and above were considered clinically significant PCa for the scope of this study. Histological results of prostate biopsy cores were anatomically mapped to corresponding mpMRI ROIs using biopsy plans. T1 relaxation times and ADC values were compared across prostate regions and ISUP groups. Across different strata, T1 relaxation time, ADC values, and diagnostic accuracy (area under the curve [AUC]) were compared using statistical methods accounting for clustered data.

RESULTS: Of 67 eligible participants, a total of 40 participants undergoing ultrasound/MRI fusion-guided prostate biopsy were included. Multislice T1 mapping was successfully performed in all participants at a median acquisition time of 2:10 minutes without evident image artifacts. A total of 71 prostate lesions was radiologically identified (TZ 49; PZ 22). Among those, 22 were histologically diagnosed with PCa (ISUP groups 1/2/3/4 in n = 3/15/3/1 cases, respectively). In the TZ, T1 relaxation time was statistically significantly lower for PCa compared with reference regions ( P = 0.029) and benign prostate hyperplasia nodules ( P < 0.001). Similarly, in the PZ, PCa demonstrated shorter T1 relaxation times versus reference regions ( P < 0.001). PCa also showed a trend toward shorter T1 relaxation times (median, 1.40 seconds) compared with radiologically suspicious lesions with benign histology (median, 1.47 seconds), although statistical significance was not reached ( P = 0.066). For discrimination of PCa from reference regions and benign prostate lesions, T1 relaxation times and ADC values demonstrated AUC = 0.80 and AUC = 0.83, respectively ( P = 0.519). Discriminating PCa from radiologically suspicious lesions with benign histology, T1 relaxation times and ADC values showed AUC = 0.69 and AUC = 0.62, respectively ( P = 0.446).

CONCLUSIONS: T1FLASH-based T1 mapping yields robust results for quantification of prostate T1 relaxation time at a short examination time of 2:10 minutes without evident image artifacts. Associated T1 relaxation times could aid in discrimination of significant and nonsignificant PCa. Further studies are warranted to confirm these results in a larger patient cohort, to assess the additional benefit of T1FLASH maps in conjunction with mpMRI sequences in the setting of deep learning, and to evaluate the robustness of T1FLASH maps compared with potentially artifact-prone diffusion-weighted imaging sequences.

来源:Invest Radiol,查看原文。版权归原作者所有。

引用

  1. Al-Bourini, Seif Amir Hosseini, Giganti, et al. T1 Mapping of the Prostate Using Single-Shot T1FLASH: A Clinical Feasibility Study to Optimize Prostate Cancer Assessment. Invest Radiol. 2022. PMID: 36729865. doi:10.1097/RLI.0000000000000945

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。