669 Background: Neoadjuvant chemotherapy provides an excellent opportunity for objective assessment of treatment-induced tumor response and for studying biomarkers characteristic of therapy-induced tumor responses. Methods: Proteomic analysis of T3/T4 breast cancer was performed in patients with locally advanced breast cancer in a phase II clinical trial. The breast cancer specimen was obtained before and after four cycles of Taxotere/Carboplatin/±Herceptin treatment. Two proteomic approaches, SELDI mass spectrometry and Clontech Ab Microarray 500, were used to screen for protein biomarkers that predict response of breast cancer to chemotherapy. Results: Five tumors with pathologically complete response (pCR) and 29 tumors with various amounts of residual tumors (Non-pCR) were analyzed by SELDI-TOF using the NP 20 chip. The normalized mass signals were compared between pCR vs Non-pCR at each aligned location by Wilcoxon rank sum test. Statistically significant differences were found at 22 m/z locations using a liberal p <0.20 criterion. The best univariate predictor occurred at m/z 14960 (p=0.004), which correctly classified 5/5 pCR spectra (100%) and 24/29 Non-pCR spectra (83%). A multivariate classification tree developed using m/z 14960 and m/z 12138 intensities correctly classified all 34 spectra. Ab microarray analysis was performed on five pCR tumors and in five tumors with the largest residual cancer. The Internal Normalization Ratio (INR) was calculated and used to compare the difference of protein expression between the two groups. Eight differentially expressed protein biomarkers were selected with the criteria of a statistically significant (Student t, p<0.05) expression change of <0.77 or >1.3 fold. Three proteins (Tat-SF1, PYK2 and PTP1B) were higher, and five (E2F2, IL1b, FEN1, CDC37 and ACM1) were lower in tumors with pCR. The unsupervised hierarchical clustering of the 10 samples by these eight proteins completely separated the pCR tumors from the poor responders. Conclusions: Our study suggests that bothSELDImass spectrometry and antibody microarray may be used to predict the tumor response to neoadjuvant chemotherapy. Proteomic analysis may be useful in developing tailored chemotherapy for breast cancer. [Table: see text]
A retrospective study of prognostic factors in locally advanced cancer of uterine cervix in a Rural Medical College of West Bengal
