文献速递泌尿生殖腹部MR

患者相关特征可预测PI-RADS 4-5男性中的前列腺癌,以进一步优化MRI诊断性能

Patient-related characteristics predict prostate cancers in men with PI-RADS 4-5 to further optimize the diagnostic performance of MRI.
PubMed 文献 · Abdom Radiol (NY) · 2023-09-30

研究纳入833例接受mp-MRI和前列腺活检、PI-RADS 4-5的患者,用病变位置、年龄、PSA、fPSA、fPSA/PSA、PSAD和BMI建立预测模型。PI-RADS 4亚组模型含病变位置、年龄、fPSA/PSA和PSAD,AUC 0.748;PI-RADS 5亚组模型含PSA和PSAD,AUC 0.893。作者认为基于所选特征建立的预测模型对PI-RADS 4-5患者的前列腺癌可达到中等风险预测。

要点

  • PI-RADS 4亚组模型纳入病变位置、年龄、fPSA/PSA和PSAD,AUC 0.748(95%CI 0.694-0.803)。
  • PI-RADS 5亚组模型仅用PSA和PSAD,AUC 0.893(95%CI 0.844-0.941)。
  • 两个模型均基于患者相关临床特征,用于优化MRI活检前风险分层。

核心结论

  • 患者相关临床特征是前列腺癌的显著预测因子,基于所选特征建立的预测模型对PI-RADS 4-5患者的前列腺癌可达到中等风险预测。

研究设计与样本

  • 预测模型研究,纳入2014年2月至2022年10月期间接受mp-MRI和前列腺活检的患者833例;以活检结果为参考标准,采用单因素和多因素logistic回归筛选前列腺癌独立预测因子并建立预测模型。

诊断标准与关键数值

  • PI-RADS 4亚组预测模型含病变位置、年龄、fPSA/PSA和PSAD,AUC 0.748(95%CI 0.694-0.803),敏感度61.88%,特异度85.32%,PPV 92.52%,NPV 43.26%。
  • PI-RADS 5亚组预测模型含PSA和PSAD,AUC 0.893(95%CI 0.844-0.941),敏感度81.40%,特异度84.85%,PPV 98.37%,NPV 28.87%。

临床应用与阅片要点

  • 对PI-RADS 4患者,可结合病变位置、年龄、fPSA/PSA和PSAD评估前列腺癌风险。
  • 对PI-RADS 5患者,可结合PSA和PSAD评估前列腺癌风险。

局限性

  • 原文未报告局限性。

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: To develop a prediction model based on patient-related characteristics for detecting prostate cancer (PCa) in patients with Prostate Imaging Reporting and Data System (PI-RADS) 4-5 in multiparametric magnetic resonance imaging (mp-MRI), aiming to optimize pre-biopsy risk stratification in MRI.

MATERIALS AND METHODS: The patient-related characteristics including the lesion location, age, prostate-specific antigen (PSA), free prostate-specific antigen (fPSA), fPSA/PSA, prostate-specific antigen density (PSAD) and body mass index (BMI) were collected for patients who underwent mp-MRI and prostate biopsy between February 2014 and October 2022. Univariate and multivariate logistic regression analyses were conducted to select independent predictors of PCa and further create a prediction model. The diagnostic performance was evaluated using the area under the receiver operating characteristic curve (AUC). Moreover, sensitivity, specificity, positive-predictive value (PPV) and negative-predictive value (NPV) were also calculated.

RESULTS: A total of 833 patients were included in this study. In the subgroup PI-RADS 4, the independent characteristics of lesion location, age, fPSA/PSA and PSAD were selected to create the prediction model with an AUC of 0.748 (95% CI 0.694-0.803), sensitivity of 61.88%, specificity of 85.32%, PPV of 92.52%, and NPV of 43.26%. Besides, the prediction model in PI-RADS 5 was created using PSA and PSAD with an AUC of 0.893 (95% CI 0.844-0.941), sensitivity of 81.40%, specificity of 84.85%, PPV of 98.37% and NPV of 28.87%.

CONCLUSION: The patient-related clinical characteristics were significant predictors of PCa and the prediction model based on selected characteristics could achieve a medium risk prediction of PCa in PI-RADS 4-5.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Xiang L, Ma S, Xu Y, et al. Patient-related characteristics predict prostate cancers in men with PI-RADS 4-5 to further optimize the diagnostic performance of MRI. Abdom Radiol (NY). 2023. PMID: 37776336. doi:10.1007/s00261-023-04011-y

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。