Research Of Diagnosis For Primary Hepatic Carcinoma By SELDI Protein Chip Technology

Objective Using SELDI Protein chip technology to detect the serum proteins spectrum of Primary hepatic carcinoma(PHC) with AFP-negative or low AFP level, Chronic liver disease(CLD)and healthy people, and to establish diagnosis models which can improve the diagnosis efficiency of PHC in combination with sera AFP .Methods The Surface enhanced laser desorption/ionization time-of-flight mass spectrometry(SELDI-TOF-MS) and CM10 ProteinChip technology were used to analyze the serum proteins spectrum in 33 cases of AFP-negative PHC, 35 cases of healthy people, 30 cases of AFP-negative CLD, 28 cases of low AFP level PHC and 34 cases of low AFP level CLD. Differential expressed proteins peaks were screened and diagnosis models were established by Ciphergen ProteinChip Software 3.2 and Biomarker Patterns Software 5.0.Results1.Differential expressed proteins peaks in the sera were screened: 32 differential expressed protein peaks were screened in the AFP-negative PHC group compared with those in the healthy adult group(P<0.05), with 6 up-regulated and 26 down-regulated. 7 differential peaks were screened in the AFP-negative PHC group compared with those in the AFP-negative CLD group (P<0.05), with 3 up-regulated and 4 down-regulated. 29 differential peaks were screened in the low AFP level PHC group compared with those in the low AFP level CLD group (P< 0.05), with 5 up-regulated and 24 down-regulated.2.Diagnosis models were successfully established: Diagnosis model 1 which combined with 4 candidate proteins with the m/z value of 4927.88, 2136.29, 3268.69, 3509.16 can predict AFP-negative PHC with an accuracy of 80.9%(55/68), a sensitivity of 81.8%(27/33), a specificity of 80.0%(28/35), a positive predictive value of 79.4%(27/34), a negative predictive value of 82.4%(28/34), a positive likelihood ratio of 4.091, and a negative likelihood ratio of 0.227. Diagnosis model 2 which combined with 3 candidate proteins with the m/z value of 4654.68,4967.66,3264.62 can predict AFP-negative PHC with an accuracy of 82.5%(52/63), a sensitivity of 84.8% (28/33), a specificity of 80.0% (24/30), a positive predictive value of 82.4%(28/34), a negative predictive value of 82.8%(24/29), a positive likelihood ratio of 4.242, and a negative likelihood ratio of 0.189. Diagnosis model 3 which established by candidate protein with the m/z value of 6451.62 can predict low AFP level PHC with an accuracy of 85.5%(53/62), a sensitivity of 82.1% (23/28), a specificity of 88.2% (30/34), a positive predictive value of 85.2%(23/27), a negative predictive value of 85.7%(30/35), a positive likelihood ratio of 6.982, and a negative likelihood ratio of 0.202.Conclusion1. We have successfully screened out the differential peaks in the sera which may be the associated protein, peptid or metabolites in the development and progression of hepatocellular carcinoma, the research maybe contribute to reveal the pathogenesis of PHC. 2. The diagnosis models established by SELDI-TOF-MS can improve the diagnosis efficiency of PHC in combination with sera AFP .