Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, Ohio, USA.
Robert J. Tomisch Pathology and Laboratory Medicine Institute, Cleveland Clinic, Ohio, USA.
Br J Radiol. 2023 Mar 1;96(1144):20220663. doi: 10.1259/bjr.20220663. Epub 2023 Feb 20.
Pelvic lymph node metastasis (PLNM) at the time of radical prostatectomy (RP) portends an unfavorable prognosis in prostate cancer patients. Conventional and advanced imaging remains limited in its ability to detect PLNM. We sought to evaluate the combination of a genomic classifier Decipher with Prostate Imaging Reporting and Data System (PI-RADS) scores in improving the detection of PLNM.
A retrospective review was performed of patients whom underwent RP, Decipher analysis, and pre-operative prostate MRI. Categorical variables were compared using Pearson chi-squareχ2 tests. Quantitative variables were assessed with Wilcoxon rank-sum tests. Multivariable logistic regression was used to identify predictors of PLNM on final pathology.
In total, 202 patients were included in the analysis, 23 of whom (11%) had PLNM. Patients with PLNM had higher median Decipher scores (0.73) than those without PLNM (0.61; p = 0.003). Patients with PLNM were more likely to demonstrate PI-RADS scores ≥ 4 (96%) than those without PLNM (74%; p = 0.012). Logistic regression demonstrated an interaction between Decipher score with PI-RADS score ≥4 (OR = 20.41; 95% CI, 2.10-198.74; p = 0.009) The combination demonstrated an area under the curve (AUC) of 0.73 (95% CI, 0.63-0.82; p < 0.001) for predicting PLNM.
The combination of elevated Decipher genomic score (≥ 0.6) and clinically significant PI-RADS score (≥ 4) is associated with PLNM at the time of RP in a modern high-risk cohort of patients with PCaprostate cancer.
Prostate MRI and genomic testing may help identify patients with adverse pathology.
根治性前列腺切除术（RP）时的盆腔淋巴结转移（PLNM）预示着前列腺癌患者预后不良。传统和先进的影像学在检测 PLNM 方面仍存在局限性。我们试图评估基因组分类器 Decipher 与前列腺影像报告和数据系统（PI-RADS）评分的结合，以提高 PLNM 的检测能力。
对接受 RP、Decipher 分析和术前前列腺 MRI 的患者进行回顾性研究。使用 Pearson chi-square χ2 检验比较分类变量。使用 Wilcoxon 秩和检验评估定量变量。使用多变量逻辑回归来确定最终病理上 PLNM 的预测因素。
共纳入 202 例患者进行分析，其中 23 例（11%）存在 PLNM。PLNM 患者的中位数 Decipher 评分（0.73）高于无 PLNM 患者（0.61；p = 0.003）。PLNM 患者更有可能表现出 PI-RADS 评分≥4（96%），而非 PLNM 患者（74%；p = 0.012）。逻辑回归显示 Decipher 评分与 PI-RADS 评分≥4 之间存在交互作用（OR = 20.41；95%CI，2.10-198.74；p = 0.009）。该组合对预测 PLNM 的曲线下面积（AUC）为 0.73（95%CI，0.63-0.82；p<0.001）。
在现代高危前列腺癌患者队列中，升高的 Decipher 基因组评分（≥0.6）和具有临床意义的 PI-RADS 评分（≥4）的组合与 RP 时的 PLNM 相关。
前列腺 MRI 和基因组检测可能有助于识别具有不良病理的患者。
