Gan Jonathan M, Kikano Elias G, Smith Daniel A, Rao Sanjay, Podury Ruchika, Wang Margaret, Durieux Jared C, Paspulati Raj M, Ponsky Lee, Ramaiya Nikhil H, Tirumani Sree H
Department of Radiology, University Hospitals Cleveland Medical Center, Case Western Reserve University, 11100 Euclid Ave, B114, Cleveland, OH 44106.
Center for Clinical Research, University Hospitals Cleveland Medical Center, Case Western Reserve University, Cleveland, OH.
AJR Am J Roentgenol. 2022 May;218(5):859-866. doi: 10.2214/AJR.21.26569. Epub 2021 Nov 24.
The frequency of clinically significant prostate cancer (csPCa) following negative biparametric MRI (bpMRI) and multiparametric MRI (mpMRI) has not been well investigated in direct comparative studies. The purposes of this study were to compare the frequency of csPCa after negative prebiopsy bpMRI and mpMRI and to evaluate factors predictive of csPCa in the two cohorts. This retrospective study included 232 men (mean age, 64.5 years) with negative bpMRI from August 2017 to March 2020 and 193 men (mean age, 69.0 years) with negative mpMRI from January 2018 to December 2018. PI-RADS category 1 or 2 was defined as negative. The study institution offered bpMRI as a low-cost self-pay option for patients without insurer coverage of prebiospy mpMRI. Patient characteristics and subsequent biopsy results were recorded. CsPCa was defined as Gleason score of 3 + 4 or greater. Multivariable regression analyses were performed to identify independent predictors of csPCa. The AUC of PSA density (PSAD) for csPCA was computed, and the diagnostic performance of PSAD was assessed at a clinically established threshold of 0.15 ng/mL. Systematic biopsy was performed after negative bpMRI for 41.4% (96/232) of patients and after negative mpMRI for 30.5% (59/193) ( = .02). Among those undergoing biopsy, csPCa was present in 15.6% (15/96) in the bpMRI cohort versus 13.6% (8/59) in the mpMRI cohort ( = .69). The NPV for csPCa was 84% (81/96) for bpMRI and 86% (51/59) for mpMRI. In multivariable analyses, independent predictors of csPCa included smaller prostate volume (OR, 0.27; < .001) and greater PSAD (OR, 3.09; < .001). In multivariable models, bpMRI (compared with mpMRI) was not independently predictive of csPCa ( > .05). PSAD had an AUC for csPCa of 0.71 (95% CI, 0.56-0.87) in the bpMRI cohort versus 0.68 (95% CI, 0.42-0.93) in the mpMRI cohort. For detecting csPCa, a PSAD threshold of 0.15 ng/mL had NPV of 90% and PPV of 28%, in the bpMRI cohort versus NPV of 92% and PPV of 44% in the mpMRI cohort. The frequencies of csPCa were not significantly different at systematic biopsy performed after negative bpMRI and mpMRI examinations. PSAD had similar diagnostic utility for csPCa in the two cohorts. Either bpMRI or mpMRI, in combination with PSAD measurement, can help avoid negative prostate biopsies.
在直接比较研究中，对双参数MRI（bpMRI）和多参数MRI（mpMRI）结果为阴性后临床显著前列腺癌（csPCa）的发生率尚未进行充分研究。本研究的目的是比较活检前bpMRI和mpMRI结果为阴性后csPCa的发生率，并评估两个队列中csPCa的预测因素。这项回顾性研究纳入了2017年8月至2020年3月期间bpMRI结果为阴性的232名男性（平均年龄64.5岁）以及2018年1月至2018年12月期间mpMRI结果为阴性的193名男性（平均年龄69.0岁）。PI-RADS 1类或2类被定义为阴性。该研究机构为没有保险覆盖活检前mpMRI的患者提供bpMRI作为低成本自费选项。记录患者特征和后续活检结果。CsPCa定义为Gleason评分3 + 4或更高。进行多变量回归分析以确定csPCa的独立预测因素。计算csPCA的PSA密度（PSAD）的AUC，并在临床确定的0.15 ng/mL阈值下评估PSAD的诊断性能。41.4%（96/232）的患者在bpMRI结果为阴性后进行了系统活检，30.5%（59/193）的患者在mpMRI结果为阴性后进行了系统活检（P = .02）。在接受活检的患者中，bpMRI队列中csPCa的发生率为15.6%（15/96），而mpMRI队列中为13.6%（8/59）（P = .69）。bpMRI对csPCa的阴性预测值为84%（81/96），mpMRI为86%（51/59）。在多变量分析中，csPCa的独立预测因素包括较小的前列腺体积（OR，0.27；P < .001）和较高的PSAD（OR，3.09；P < .001）。在多变量模型中，bpMRI（与mpMRI相比）不是csPCa的独立预测因素（P > .05）。PSAD在bpMRI队列中对csPCa的AUC为0.71（95% CI，0.56 - 0.87），在mpMRI队列中为0.68（95% CI，0.42 - 0.93）。对于检测csPCa，在bpMRI队列中，PSAD阈值为0.15 ng/mL时，阴性预测值为90%，阳性预测值为28%，而在mpMRI队列中，阴性预测值为92%，阳性预测值为44%。在bpMRI和mpMRI检查结果为阴性后进行的系统活检中，csPCa的发生率没有显著差异。PSAD在两个队列中对csPCa具有相似的诊断效用。bpMRI或mpMRI与PSAD测量相结合，有助于避免前列腺活检阴性。
