Validation of Biomarkers of the Tumor Microenvironment

Abstract

Currently the diagnosis of prostate cancer rests on the results of a biopsy. There are over one million biopsies performed in the U.S. alone every year. Most, about 63%, are negative. Several large series have revealed that 30-40% of these miss tumors in spite of taking 12 or more cores under ultrasound guidance. Patients recommended for repeat biopsy usually in 3-12 months receive no treatment while tumor may progress. However all biopsies contain ample stroma. Others and we have observed hundreds of gene activity changes in stroma near tumor. We have developed and published a general classifier for the diagnosis of prostate cancer based on an RNA profile of 131 genes derived from analysis of frozen prostatectomy and normal stroma samples. It has been tested on 364 independent cases and shown to be 97% accurate. Preliminary studies indicate that equivalent performance may be obtained from approximately 45-65 genes with the highest differential expression.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 2015
Accession Number
AD1007012

Entities

People

  • Dan Mercola

Organizations

  • University of California

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Anatomy
  • Biological Markers
  • Biological Sciences
  • Biomedical Research
  • California
  • Department Of Defense
  • Gene Expression
  • Machine Learning
  • Medical Personnel
  • Neoplasms
  • Prostate
  • Prostate Cancer
  • Retraining
  • Tissues
  • Training
  • Validation

Readers

  • Immunology
  • Medical Imaging.
  • Oncology