Epithelium percentage estimation facilitates epithelial quantitative protein measurement in tissue specimens.

BACKGROUND: The rapid advancement of high-throughput tools for quantitative measurement of proteins has demonstrated the potential for the identification of proteins associated with cancer. However, the quantitative results on cancer tissue specimens are usually confounded by tissue heterogeneity, e.g. regions with cancer usually have significantly higher epithelium content yet lower stromal content. OBJECTIVE: It is therefore necessary to develop a tool to facilitate the interpretation of the results of protein measurements in tissue specimens. METHODS: Epithelial cell adhesion molecule (EpCAM) and cathepsin L (CTSL) are two epithelial proteins whose expressions in normal and tumorous prostate tissues were confirmed by measuring staining intensity with immunohistochemical staining (IHC). The expressions of these proteins were measured by ELISA in protein extracts from OCT embedded frozen prostate tissues. To eliminate the influence of tissue heterogeneity on epithelial protein quantification measured by ELISA, a color-based segmentation method was developed in-house for estimation of epithelium content using H&E histology slides from the same prostate tissues and the estimated epithelium percentage was used to normalize the ELISA results. The epithelium contents of the same slides were also estimated by a pathologist and used to normalize the ELISA results. The computer based results were compared with the pathologist's reading. RESULTS: We found that both EpCAM and CTSL levels, measured by ELISA assays itself, were greatly affected by epithelium content in the tissue specimens. Without adjusting for epithelium percentage, both EpCAM and CTSL levels appeared significantly higher in tumor tissues than normal tissues with a p value less than 0.001. However, after normalization by the epithelium percentage, ELISA measurements of both EpCAM and CTSL were in agreement with IHC staining results, showing a significant increase only in EpCAM with no difference in CTSL expression in cancer tissues. These results were obtained with normalization by both the computer estimated and pathologist estimated epithelium percentage. CONCLUSIONS: Our results show that estimation of tissue epithelium percentage using our color-based segmentation method correlates well with pathologists' estimation of tissue epithelium percentages. The epithelium contents estimated by color-based segmentation may be useful in immuno-based analysis or clinical proteomic analysis of tumor proteins. The codes used for epithelium estimation as well as the micrographs with estimated epithelium content are available online.

Identification, prioritization, and evaluation of glycoproteins for aggressive prostate cancer using quantitative glycoproteomics and antibody-based assays on tissue specimens.

Prostate cancer is highly heterogeneous in nature; while the majority of cases are clinically insignificant, some cases are lethal. Currently, there are no reliable screening methods for aggressive prostate cancer. Since most established serum and urine biomarkers are glycoproteins secreted or leaked from the diseased tissue, the current study seeks to identify glycoprotein markers specific to aggressive prostate cancer using tissue specimens. With LC-MS/MS glycoproteomic analysis, we identified 350 glycopeptides with 17 being altered in aggressive prostate cancer. ELISA assays were developed/purchased to evaluate four candidates, that is, cartilage oligomeric matrix protein (COMP), periostin, membrane primary amine oxidase (VAP-1), and cathepsin L, in independent tissue sets. In agreement with the proteomic analysis, we found that COMP and periostin expressions were significantly increased in aggressive prostate tumors while VAP-1 expression was significantly decreased in aggressive tumor. In addition, the expression of these proteins in prostate metastases also follows the same pattern observed in the proteomic analysis. This study provides a workflow for biomarker discovery, prioritization, and evaluation of aggressive prostate cancer markers using tissue specimens. Our data suggest that increase in COMP and periostin and decrease in VAP-1 expression in the prostate may be associated with aggressive prostate cancer.

Longitudinal analysis of pain in patients with metastatic prostate cancer using natural language processing of medical record text.

OBJECTIVES: To test the feasibility of using text mining to depict meaningfully the experience of pain in patients with metastatic prostate cancer, to identify novel pain phenotypes, and to propose methods for longitudinal visualization of pain status. MATERIALS AND METHODS: Text from 4409 clinical encounters for 33 men enrolled in a 15-year longitudinal clinical/molecular autopsy study of metastatic prostate cancer (Project to ELIminate lethal CANcer) was subjected to natural language processing (NLP) using Unified Medical Language System-based terms. A four-tiered pain scale was developed, and logistic regression analysis identified factors that correlated with experience of severe pain during each month. RESULTS: NLP identified 6387 pain and 13 827 drug mentions in the text. Graphical displays revealed the pain 'landscape' described in the textual records and confirmed dramatically increasing levels of pain in the last years of life in all but two patients, all of whom died from metastatic cancer. Severe pain was associated with receipt of opioids (OR=6.6, p<0.0001) and palliative radiation (OR=3.4, p=0.0002). Surprisingly, no severe or controlled pain was detected in two of 33 subjects' clinical records. Additionally, the NLP algorithm proved generalizable in an evaluation using a separate data source (889 Informatics for Integrating Biology and the Bedside (i2b2) discharge summaries). DISCUSSION: Patterns in the pain experience, undetectable without the use of NLP to mine the longitudinal clinical record, were consistent with clinical expectations, suggesting that meaningful NLP-based pain status monitoring is feasible. Findings in this initial cohort suggest that 'outlier' pain phenotypes useful for probing the molecular basis of cancer pain may exist. LIMITATIONS: The results are limited by a small cohort size and use of proprietary NLP software. CONCLUSIONS: We have established the feasibility of tracking longitudinal patterns of pain by text mining of free text clinical records. These methods may be useful for monitoring pain management and identifying novel cancer phenotypes.