Goodison Steve, Ogawa Osamu, Matsui Yoshiyuki, Kobayashi Takashi, Miyake Makito, Ohnishi Sayuri, Fujimoto Kiyohide, Dai Yunfeng, Shimizu Yoshiko, Tsukikawa Kazue, Furuya Hideki, Rosser Charles J
Department of Health Sciences Research, Mayo Clinic, Jacksonville, FL, USA.
Mayo Clinic Cancer Center, Rochester, MN, USA.
J Transl Med. 2016 Oct 7;14(1):287. doi: 10.1186/s12967-016-1043-1.
Bladder cancer (BCa) is among the most commonly diagnosed malignancies worldwide, and due the high rate of post-operative disease recurrence, it is one of the most prevalent in many countries. The development of non-invasive molecular assays that can accurately detect and monitor BCa would be a major advance, benefiting both patients and healthcare systems. We have previously identified a urinary protein biomarker panel that is being developed for application in at-risk patient cohorts. Here, we investigated the potential utility of the multiplex assay in a Japanese cohort.
The Japanese study cohort collected from urology clinics at two institutions was comprised of a total of 288 subjects. The protein biomarker panel (IL8, MMP9, MMP10, ANG, APOE, SDC1, A1AT, PAI1, CA9, VEGFA) was monitored in voided urine samples collected prior to cystoscopy using a custom multiplex ELISA assay. The diagnostic performance of the biomarker panel was assessed using receiver operator curves, predictive modeling and descriptive statistics.
Urinary biomarker concentrations were significantly elevated in cases versus controls, and in cases with high-grade and muscle-invasive tumors. The AUC for the 10-biomarker assay was 0.892 (95 % confidence interval 0.850-0.934), with an overall diagnostic sensitivity specificity of 0.85 and 0.81, respectively. A predictive model trained on the larger institutional cohort correctly identified 99 % of the cases from the second institution.
Urinary levels of a 10-biomarker panel enabled discrimination of patients with BCa. The multiplex urinary diagnostic assay has the potential to be developed for the non-invasive detection of BCa in at-risk Japanese patients.
膀胱癌(BCa)是全球最常见的诊断恶性肿瘤之一,由于术后疾病复发率高,它在许多国家也是最普遍的癌症之一。能够准确检测和监测膀胱癌的非侵入性分子检测方法的开发将是一项重大进展,这对患者和医疗系统都有益。我们之前已经鉴定出一种尿液蛋白质生物标志物组合,目前正在开发将其应用于高危患者队列。在此,我们研究了该多重检测方法在日本队列中的潜在效用。
从两家机构的泌尿外科诊所收集的日本研究队列共包括288名受试者。使用定制的多重ELISA检测法,对膀胱镜检查前收集的晨尿样本中的蛋白质生物标志物组合(IL8、MMP9、MMP10、ANG、APOE、SDC1、A1AT、PAI1、CA9、VEGFA)进行监测。使用受试者操作曲线、预测模型和描述性统计来评估生物标志物组合的诊断性能。
与对照组相比,病例组以及高级别和肌肉浸润性肿瘤病例组的尿液生物标志物浓度显著升高。10种生物标志物检测的曲线下面积(AUC)为0.892(95%置信区间0.850 - 0.934),总体诊断敏感性和特异性分别为0.85和0.81。在较大机构队列上训练的预测模型正确识别了来自第二家机构的99%的病例。
10种生物标志物组合的尿液水平能够区分膀胱癌患者。这种多重尿液诊断检测方法有潜力被开发用于对日本高危患者进行膀胱癌的非侵入性检测。