Health & Environmental Research Online (HERO)


Print Feedback Export to File
2571306 
Journal Article 
A serum protein fingerprint in the diagnosis and prognosis of Wilms' tumors in children 
Zhang, Bo; Wang, J; Yu, J; Zheng, Shu 
2009 
1 
Life Science Journal
ISSN: 1097-8135 
6 
1 
27-32 
WOS:000265884000005 
Objective. To establish specific fingerprints of serum markers for the diagnosis and prognosis of childhood Wilms' tumors. Method. A total of 120 serum samples include 30 samples of Wilms' tumor patients after diagnosis, 70 samples collected after nephrectomy and 20 samples collected from healthy children. The sera were analyzed by Ciphergen ProteinChip coupled with surface-enhanced laser desorption ionization time-of-flight mass spectrometry (SE LDI-TOF-MS) to characterize the mass spectrometric profiles. The data were then classified by support vector machine (SVM) analyses. Results. A protein with mass/charge (M/Z) ratio of 6984.5 (pM/Z6984.5) and a protein with M/Z ratio of 6455.5 (pM/Z6455.5) were identified based on their progressive down-regulation in Wilms' tumor. In addition, a protein with M/Z ratio of 4153.9 (pM/Z4153.9) and a protein with M/Z ratio of 3257.6 (pM/Z3257.6) were found to be specifically associated with the transition from Wilms' tumor from stage I/II to stage III/IV. These data were cross-validated using the leave-one out method. The accuracy of clinical staging afforded by the protein chip technique was 100%. This result was consistent with pathological staging and was superior to computed tomography (CT) scan. Both pM/Z6984.5 and pM/Z6455.5 were weakly not expressed in pre-surgical serum specimens. However, their levels were elevated in 19 of 21 samples from patients who just underwent radical resection to a level comparable as found in normal controls. Interestingly, the expression of pM/Z6984.5 and pM/Z6455.5 in five cases remained low even at three and at six months after nephrectomies. Conclusion. The combination of SELDI-TOF-MS and SVM led to serum protein fingerprint comprised of pM/Z6984.5, pM/Z6455, pM/Z4153.9 and pM/Z3257.6 for Wilms' tumor. This fingerprint model may prove to be highly specific and sensitive for the diagnosis and prognosis. [Life Science Journal. 2009; 6(1): 27 - 32] (ISSN: 1097 - 8135). 
SELDI-TOF-MS; support vector machine; tumor stage; Wills' tumor