الفهرس 
I. IntroductionOnly 14 pages are availabe for public view
 |  | from | 146 |  |  |
| | | from | 146 | | |
Abstract
HCC is the most common primary malignancy of the liver, accounting for about 80% of all liver cancers and occurs predominantly in patients with underlying chronic liver disease and cirrhosis. HCC is increasing in incidence and has a very high fatality rate. It is the fifth most common malignancy in male, the seventh in female and the third leading cause of cancer-associated deaths around the world, and more than 800,000 deaths are reported internationally each year.
Viral hepatitis plays a major role in the development of HCC. Hepatitis C virus causes acute and chronic hepatitis and can lead to permanent liver damage and HCC via oxidative stress, insulin resistance, fibrosis, liver cirrhosis, steatosis and steato-hepatitis.
Treatment is effective in the early stages of HCC, necessitating the early diagnosis which until now remains difficult. The current methods for early detection of HCC rely on several tools combining imaging techniques and serologic markers. Therefore, sensitive and specific serum markers capable of detecting HCC at an early stage are highly required.
Although AFP is a widely used serological marker for detection of HCC, its sensitivity and specificity are not optimal. AFP is not secreted in all cases of HCC and may be normal in as many as 40% of patients with early HCC. Furthermore, the serum level can be elevated in cirrhosis as well as hepatitis without associated HCC. To improve the diagnosis and prognosis of HCC, additional reliable markers must be identified that can be used for its early and accurate detection.
Proteomics, which concerns comprehensive protein profile changes caused by multiple gene alterations, has emerged as a valuable tool in scientific medicine. Proteomic pattern by mass spectrometry is one of the most promising new approaches for early identification of potential biomarker, because various proteins are expressed and released into the bloodstream in response to specific physiological and pathological states.
Among the analytical platforms that assess the proteome, MALDI TOF- mass spectrometry is one of the most commonly used approaches for proteomic studies, because it provides good separation and accurate detection of proteins in complex specimens with high sensitivity and resolution. Magnetic beads with different chemical chromatographic surfaces selectively bind proteins which are eluted and analyzed with MALDI-TOF MS.
In the current study, serum from HCV related 60 LC patients without HCC, 60 LC patients with HCC and 60 healthy volunteers were collected. The three groups were age matched. For proteomic analysis, Magsi- WCX MB which bind proteins of the mass range 1-20 KDa, ultraflextreme MALDI-TOF MS and ClinProTools were used. For profile generation, each group was divided into two independent sets randomly: a training set and validation set. Genetic Algorithm (GA) was used to set up the classification model which represented the proteomic profile.
In HCC, male predominance is more obvious in population at high risk with male to female ratio 3.7:1. This observation was confirmed by our results. The M/F ratios were 3/1 and 4/1 in LC and HCC patients respectively. Besides the higher ratio of males in HCV infection, the male predominance in HCC may be explained by greater exposure of males to other environmental carcinogens, a role for sex hormones, and higher DNA synthetic activities in male cirrhotic patients compared to females.
According to Child–Pugh classification in the current study, in LC patients, Child A represent 5%, Child B 48.3% and Child C 46.7% of cases. While in HCC patients, Child A represents 60%, Child B 38.3% and Child C 1.7% of the cases. According to BCLC classification 81.7% of HCC cases were in stage A, while 17.3% were in stage B.
Proteomic maps associated with the disease were drawn and the total number of peaks were 104, 108, 123 with 47, 73 and 70 differential significant peptide peaks respectively appeared in the three groups (first group: LC versus controls) (second group: HCC versus controls) (third group: HCC versus LC).
A profile of 5 peaks was generated by GA to differentiate between HCC and healthy controls at m/z 1520.39, 2107.26, 5804.98, 4128.57and 8141.25 with recognition capacity of 99.4% and an overall cross validation of 92.4%. External validation of this profile using the validation sets yield sensitivity of 88.4 % and specificity of 84.5 %. This may be defined as the leading differential peptides associated with HCC, which are worthy of further sequence determination and functional analysis.
A profile of 5 peaks was generated by GA to differentiate between LC and healthy controls at m/z 2754.25, 4070.32, 1920.87, 2555.67, 3928.09 with recognition capacity of 96.2 % and an overall cross validation of 84.8 %. External validation of this profile using the validation sets yield sensitivity of 83 % and specificity of 82.2 %.
Another profile of 5 peaks was generated by GA to differentiate between HCC and LC patients at m/z 941.66, 2021.52, 4246.54, 6271.55 and 6463.58 with recognition capacity of 93.6 % and an overall cross validation of 82.7 %. External validation of this profile using the validation sets yield sensitivity of 77.2% and specificity of 69.2%.
Serum AFP level was > 10.0 ng/mL in 18 patients with LC (30%), on the other hand 50% of HCC cases had AFP level less than 20 ng/ml which means that around half of the cases, AFP level was near normal, AFP level >400 ng/ml was found in16.6% of the cases, and >100 in 25% and > 200 in 16.6%.
MALDI-TOF profiling was used, also to classify HCC patients according to AFP level, for patients with AFP < 100 ng/ml and those with AFP > 100 ng/ml. Using GA, a profile of 5 peaks at m/z 931.59, 8308.54, 5098.34, 2021.05, 6998.93 was used to differentiate between the two groups with recognition capacity of 96.3% and an overall cross validation of 87.9 %.
A another Proteomic profile of 5 peaks at m/z 5063.24, 2970.45, 12604.55, 4070.5, 5289.67 was generated with GA can discriminate between HCC patients according to Child A & B class with recognition capacity 100 % and over all cross validation 89.7 %.
In the current study Peak 4070 was repeated in group I, group II, group III and in Child Pugh Class, this peak was down regulated in all groups, this can be explained by the fact that peak 4070 was under expressed in liver cirrhosis and decrease more in early HCC. While peak 5804 was repeated in group II and group III, this peak was up regulated in both groups, meaning that peak 5804 was overexpressed in HCC patients in comparison to control and in comparison to LC patients. Also peak 2021 was repeated in group III and AFP classification, this peak was down regulated in both groups. A similar peaks were reported in Liu T et al and Ying X et al studies.
Tandem mass spectrometry was not performed for the identification of the peptides of the profile discovered. However, we searched for their identity by comparing them to the literature published regarding the identity of mass spectrometry peaks reported in similar work.
Similarly Ying X et al study (2013), in their study on 32 patients with HCC (16 patients with chronic hepatitis, 16 patients with LC and 16 healthy volunteers) detected proteomic patterns of 11 peaks. Interestingly enough peak 5804.98 in our study was similar to their peak 5805.51which was also up-regulated. Also peak 4070.32 was close to their peak 4054.21, both were down regulated.
In Liu Y et al study, one of the up-regulated peak (8567-Da) in HCC sera was a fragment of apolipoprotein A-I (apo A-I).They concluded that: MALDI-TOF MS analysis revealed that apo A-I is a potential novel serum marker of HCC. Interestingly, peak 8565.07 was significantly up-regulated in our study in SNN model classification. The up regulated peaks 4246.54 and 2555.67 in our work were nearly close to their peaks 4293.88 and 2660.62, which were up regulated also. These studies suggest that a proteomic method can be used for identifying serum biomarkers for HCC and other liver diseases and also for distinguishing the HCC plus HCV-positive cirrhosis from HCV-negative cirrhosis.
As HCC is raised through multiple risk factors, it is hard to characterize personal HCC using only one single biomarker. The investigation of biomarker combinations might provide more accurate and valuable information for the future HCC diagnosis and/or prognosis. As HCC has complex and heterogeneous pathologies, combination of several biomarkers may be more relevant for diagnosis.
The results were discussed with similar results in the same field, from the previous results it is concluded that a combined application of magnetic beads with MALDI-TOF MB technique was suitable for identification of potential serum biomarkers for HCC and it is a promising way to establish a diagnostic pattern.