Department of Urology, Maastricht University Medical Centre, Maastricht, The Netherlands.
Department of Urology, Maastricht University Medical Centre, Maastricht, The Netherlands.
Eur Urol Focus. 2020 Nov 15;6(6):1220-1225. doi: 10.1016/j.euf.2018.11.006. Epub 2018 Nov 24.
Prostate biopsy, an invasive examination, is the gold standard for diagnosing prostate cancer (PCa). There is a need for a novel noninvasive diagnostic tool that achieves a significantly high pretest probability for PCa, reducing unnecessary biopsy numbers. Recent studies have shown that volatile organic compounds (VOCs) in exhaled breath can be used to detect different types of cancers via training of an artificial neural network (ANN).
To determine whether exhaled-breath analysis using a handheld electronic nose device can be used to discriminate between VOC patterns between PCa patients and healthy individuals.
DESIGN, SETTING, AND PARTICIPANTS: This prospective pilot study was conducted in the outpatient urology clinic of the Maastricht University Medical Center, the Netherlands. Patients with histologically proven PCa were already included before initial biopsy or during follow-up, with no prior treatment for their PCa. Urological patients with negative biopsies in the past year or patients with prostate enlargement (PE) with low or stable serum prostate-specific antigen were used as controls. Exhaled breath was probed from 85 patients: 32 with PCa and 53 controls (30 having negative biopsies and 23 PE).
Patient characteristics were statistically analyzed using independent sample t test and Pearson's chi-square test. Data analysis was performed by Aethena software after data compression using the TUCKER3 algorithm. ANN models were trained and evaluated using the leave-10%-out cross-validation method.
Our trained ANN showed an accuracy of 0.75, with an area under the curve of 0.79 with sensitivity and specificity of 0.84 (95% confidence interval [CI] 0.66-0.94) and 0.70 (95% CI 0.55-0.81) respectively, comparing PCa with control individuals. The negative predictive value was found to be 0.88. The main limitation is the relatively small sample size.
Our findings imply that the Aeonose allows us to discriminate between patients with untreated, histologically proven primary PCa and control patients based on exhaled-breath analysis.
We explored the possibility of exhaled-breath analysis using an electronic nose, to be used as a noninvasive tool in clinical practice, as a pretest for diagnosing prostate cancer. We found that the electronic nose was able to discriminate between prostate cancer patients and control individuals.
前列腺活检是一种有创检查，是诊断前列腺癌（PCa）的金标准。需要一种新的非侵入性诊断工具，以显著提高 PCa 的术前概率，减少不必要的活检数量。最近的研究表明，通过人工神经网络（ANN）的训练，呼气中的挥发性有机化合物（VOCs）可用于检测不同类型的癌症。
确定手持式电子鼻设备是否可用于区分 PCa 患者和健康个体之间的呼气分析中的 VOC 模式。
设计、地点和参与者：这是一项在荷兰马斯特里赫特大学医学中心的门诊泌尿科诊所进行的前瞻性试点研究。已经包括了经组织学证实患有 PCa 的患者，这些患者在初次活检前或随访期间已经进行了活检，且尚未接受过 PCa 的治疗。过去一年中前列腺活检阴性的泌尿科患者或前列腺增大（PE）且血清前列腺特异性抗原水平低或稳定的患者作为对照组。
我们的训练有素的 ANN 显示出 0.75 的准确率，曲线下面积为 0.79，灵敏度和特异性分别为 0.84（95%置信区间 [CI] 0.66-0.94）和 0.70（95% CI 0.55-0.81），将 PCa 与对照组进行比较。阴性预测值为 0.88。主要限制是样本量相对较小。
我们的研究结果表明，Aeonose 可以根据呼气分析来区分未经治疗的、组织学证实的原发性 PCa 患者和对照组患者。
我们探讨了使用电子鼻进行呼气分析的可能性，将其作为临床实践中的一种非侵入性工具，作为诊断前列腺癌的术前检查。我们发现电子鼻能够区分前列腺癌患者和对照组个体。
