Institute of Pathology, Department of Urology, RWTH Aachen University, Aachen, Germany.
Eur Urol. 2009 Dec;56(6):989-96. doi: 10.1016/j.eururo.2009.02.031. Epub 2009 Mar 6.
Despite continuing research for accurate bladder cancer biomarkers, the analytes suffer from lack of sensitivity and specificity.
To search for discriminating protein patterns in serum, we used magnetic bead-based separation followed by matrix-assisted laser desorption ionization (MALDI) time-of-flight (TOF) mass spectrometry (MS) to identify patients with bladder cancer.
DESIGN, SETTING AND PARTICIPANTS: In total, serum samples from 105 patients with bladder cancer, 98 healthy controls, and 45 prostate cancer patients were included in this study.
Serum samples were fractionated by means of surface-activated magnetic beads and were subsequently analyzed with MALDI-TOF MS. Multidimensional data analysis was done to generate algorithms capable of distinguishing between cancer patients and healthy individuals. The algorithms were trained using a training set of 41 bladder cancer patients and 39 healthy controls and were validated with an independent test set of 64 bladder cancer patients and 59 healthy controls. Additionally, 45 prostate cancer samples were used as a third test set.
In the training set, patients with bladder cancer could be identified with an overall sensitivity of 94.1% and specificity of 89.2%. Similar results could be achieved for the test set, showing 96.4% sensitivity and 86.5% specificity. Even the presence of low-stage tumors could be predicted with 96% sensitivity and could be distinguished from higher stage or grade tumors with a sensitivity of 77.3%. Distinction between other tumor stages, however, resulted in lower sensitivity values.
These findings demonstrate that screening for serum protein patterns using MALDI-TOF MS shows high sensitivity and specificity in identifying patients with bladder cancer, regardless of tumor stage. Due to high-throughput capability, the identified differential protein panel may improve the detection of bladder cancer.
尽管不断有研究寻找准确的膀胱癌生物标志物，但这些分析物仍存在灵敏度和特异性不足的问题。
为了寻找在血清中具有鉴别能力的蛋白质模式，我们使用基于磁珠的分离方法，然后结合基质辅助激光解吸电离飞行时间质谱（MALDI-TOF MS）来鉴定膀胱癌患者。
设计、地点和参与者：共纳入 105 例膀胱癌患者、98 例健康对照者和 45 例前列腺癌患者的血清样本进行本研究。
采用表面活化磁珠对血清样本进行分离，然后用 MALDI-TOF MS 进行分析。通过多维数据分析生成能够区分癌症患者和健康个体的算法。该算法使用包含 41 例膀胱癌患者和 39 例健康对照者的训练集进行训练，并使用包含 64 例膀胱癌患者和 59 例健康对照者的独立测试集进行验证。此外，还使用了 45 例前列腺癌样本作为第三个测试集。
在训练集中，膀胱癌患者的总体灵敏度为 94.1%，特异性为 89.2%。在测试集中也可以获得类似的结果，灵敏度为 96.4%，特异性为 86.5%。甚至可以预测低分期肿瘤的存在，其灵敏度为 96%，并可以将其与高分期或高分级肿瘤区分开，其灵敏度为 77.3%。然而，区分其他肿瘤分期的灵敏度值较低。
这些发现表明，使用 MALDI-TOF MS 筛选血清蛋白质模式在识别膀胱癌患者方面具有较高的灵敏度和特异性，而与肿瘤分期无关。由于高通量能力，所鉴定的差异蛋白质谱可能会提高膀胱癌的检测能力。
