Objective: To screen the distinct expression of serum proteins
of renal cancer and to identify specific serum tumor markers.
Methods: Surface-enhanced laser desorption/inionation time
of flight-mass spectrometry (SELDI-TOF-MS) and CM10
protein chip were used to detect the serum protein patterns of 72
patients with renal cancer, 30 controls with benign lesions, and
42 normal controls. The data was analyzed and the diagnostic model
was established by using ZUCI-protein chip data analysis system
software package. The data of spectra were analyzed by support
vector machine(SVM) to establish a diagnostic model, which was
further evaluated and validated by leave one cross validation.
Results: Six protein markers were identified with the relative
molecular weights of 5947, 5912, 5937, 6114, 5344, and 5984,
respectively. The differences of these protein markers between
renal cancer patients and controls were statistically significant
(P<0.01). The detective model could differentiate renal cancer
from healthy controls with a sensitivity of 91.67% (66/72) and
specificity of 93.06% (67/72). The sensitivity and specificity of
double blind confirmation procedure were 91.43% (32/35) and
92.86% (39/42), respectively.
Conclusions: The predictive models was successfully built by
the differences of serum protein fingerprint, which provides a
novel, effective, and highly specific and sensitive diagnostic tool
for renal cell carcinoma.