Debes JD et al [Cancer Res. 63, 7638-7640, 2003] demonstrated that overexpression of the P300 gene can cause alterations in nuclear structure of PCa cell lines and epithelial cells of biopsies of men with non-organ-confined PCa. We assessed the expression of the P300 protein in men with long term follow-up that had undergone radical prostatectomy (RP).

An NCI Cooperative Prostate Cancer Tissue Resource (CPCTR) tissue microarray was prepared from RP cases treated in 1991-92. The CPCTR-TMA cores from n=78 cases (n=39 non-recurrences and 39 PSA recurrences) were assayed. Feulgen-stained nuclei were captured by digital image analysis from two 0.60 mm spots for each case using the AutoCyteTM Pathology Workstation. We applied Multivariate Logistic Regression (MLR) for selection of 40 nuclear morphometric descriptors (NMD) to calculate a Quantitative Nuclear Grade (QNG) signature, routine pathology as well as P300 biomarker models.

The MLR model for QNG predicted PSA recurrence using 18/40 NMDs yielding an area under the receiver operator characteristic curve (AUC-ROC) of 0.865 with a sensitivity of 89%, specificity of 60% and an accuracy of 74%. Pathology had an AUC-ROC = 0.70 with a sensitivity of 53%, specificity of 60% and an accuracy of 56%. P300 biomarker had an AUC-ROC = 73.6 with a sensitivity of 93.5%, specificity of 28% and an accuracy of 58%. By combining QNG and the P300 variables the AUC-ROC improved marginally from 0.865 for QNG to 0.88 and specificity increased from 60% to 67%. Spearman’s rank correlation for P300 was significantly different (P < 0.05) in the PCa recurrence group for area, perimeter, feret Y and minimum feret size and shape NMD variables.

Using a TMA prepared from 78 men with 13 years follow-up, we demonstrated that the P300 biomarker predicts PCa progression and correlates with certain morphometric size and shape variables measured by image analysis. We are continuing to assess other chromatin remodeling genes modified in PCa progression which correlate to alterations in nuclear structure that are measured by QNG.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA