Research Of Serum Proteomics In Breast Cancer Based On SELDI-TOF-MS

Breast cancer is a significant threat to women health care. Hence, new markers for early diagnosis, accurate prognosis and prediction of response to treatment are warranted to improve breast cancer care. Since proteomics can bridge the gap between the genetic alterations underlying cancer and cellular physiology, much is expected from proteome analyses for the detection of better protein biomarkers. Recent technical advances in mass spectrometry have enabled high-throughput proteome analysis. This study investigated serum of breast cancer patients based on the surface-enhanced laser desorption/ionizations time of flight-mass spectrometry (SELDI-TOF-MS) technology, and it made detailed comparisons for serum proteome between breast cancers and non-cancer patients, and contrast the different expressed proteins (protein profiling) during the variant treatment stage of the breast cancer, then seeking for possible candidate biomarkers for improving cancer patient outcome of breast cancer.1. Serum protein profiling using SELDI-TOF-MS techniqueObjective: Seeking for different expressed proteins associated with different treatment stage of the breast cancer and making a preparation for further investigation. Methods: Using SELDI-TOF MS and protein chip technique to screen the serum proteins from benign breast disease and control group, and to research different expressed proteins during the variant treatment stage of the breast cancer. Result: There are 117 serum protein mass spectrometry images for preliminary diagnosed breast cancer patients, 78 (39 paired) images for patients who conceived neoadjuvant chemotherapy, and 36 (18 paired) images for patients who endured mastectomy but did not receive any former chemotherapy, 20 images for patients who encountered cancer recurrence, 85 images for benign breast disease and 56 images for healthy control group. Conclusion: Low molecular weight serum proteins can be well screened by SELDI-TOF MS technique.2. Differences of protein profiling in serum among breast cancer patients, benign breast disease patients and healthy controlObjective: To investigate the different expressed proteins in serum among breast cancer and healthy control, breast cancer and breast benign disease. Using a diagnosis model to distinguish breast cancer and non-cancer patients by screening diversity expressed proteins. Methods: Randomized divide each group into training group and testing group, using each training group to establish a optimal classification model, then validate the model with testing group by blind trial. Result: There are 17 significant deviation protein peaks among 3 groups (P<0.05), 14 peaks occurred significantly between breast cancer and healthy control group (P<0.05) and 12 peaks between breast cancer and breast benign disease group (P<0.05). 10 proteins (m/z labeled 4212, 5902, 7940, 11360, 4974, 5384, 6009, 7766, 8150 and 8518) were picked out for developing a classification model to distinguish breast cancer, healthy control and breast benign disease, In training group , the sensibility for identification of breast cancer was 85.14% (63/74), for identification of benign breast disease patients was 86.11% (31/36), and for identification of healthy women was 88% (44/50). Validating the classification model with testing group in blind trial, the sensibility in each group are 79.09%(34/43), 65.00% (13/20) and 77.14% (27/35), respectively, and the resultant accuracy is 75.51%(74/98). Conclusion: There are different protein profilings among groups, and the model set with 10 proteins which m/z labeled 4212, 5902, 7940, 11360, 4974, 5384, 6009, 7766, 8150 and 8518 can distinguish breast cancer from breast benign disease and healthy control groups effectively.3. Analyzing for serum proteome associated with neoadjuvant chemotherapy in breast cancerObjective: To investigate changes of protein profiling before and after neoadjuvant chemotherapy with paclitaxel and epirubicin (TE) program, seeking for possible therapeutic effect prediction proteins. Methods: Screening the different expressed proteins before and after neoadjuvant chemotherapy in groups with different effect, analyzing protein profiling difference in treatment effect between neoadjuvant chemotherapy groups which participated in the TE program.Result: After neochemotherapy, There are 7 protein levels decreased and 1 increased in chemotherapy effective group, and there is no significant change in inefficient group. The classification model with 5 proteins which m/z labeled 4974, 10517, 6338, 9282 and 7766 can distinguish chemotherapy effective group from ineffective group with TE program. The model may be better predictors for the effect of neoadjuvant chemotherapy in breast cancer. Conclusion: There are significant differences in serum proteome after neoadjuvant chemotherapy. Using a classification model set by proteins which m/z 4974, 10517, 6338, 9282 and 7766 can identify the patients who may be better benefit from TE program chemotherapy.4. Comprehensive analysis of breast cancer related serum proteinsObjective: To investigate the characters of different expressed proteins during the pathogenesis of breast cancer, screening for possible candidate biomarkers for breast cancer. Methods: Make comparisons for protein profiling before and after mastectomy, and contrast serum proteins between cancer recurrence and healthy control, then build a classification model, coupled with former research informations of protein profiling in variant treatment stage, seeking for possible novel candidate biomarkers for breast cancer. Result: There is only 1 protein (m/z 4974) level decreased significantly early after mastectomy operation. There are 37 significant changed protein peaks between recurrence group and healthy control, using a classification model set by protein m/z 8520 can distinguish serums from two different groups, which accuracy is 100% (50/50), sensibility is 100% (20/20) and specificity is 100% (30/30). Proteins with m/z 4973, 4285 and 7767 showed a higher protein level in breast cancer group, and a lower level both after mastectomy or chemotherapy, but a higher protein level again in recurrence patients , they together showed a positive correlation with the bearing cancer state. Conclusion: Protein m/z 8518 may be a biomarker for cancer recurrence, and m/z 4974 may be a biomarker to monitor tumor activities and body cancer bearing states, which are truly worthy of being further investigated.