NEW YORK (GenomeWeb) – A new prostate cancer analysis suggests tumors from African American men are more likely to carry gene splicing variants associated with tumor aggressiveness and poor outcomes than tumors from European Americans.
Researchers from George Washington University and elsewhere used array-based expression profiling to uncover differential splicing events in prostate cancer samples from 20 African American and 15 European American men — a search that led to thousands of candidate splice variants that were differentially expressed depending on ancestry.
The team then took a closer look at splice variants involving genes with suspected ties to cancer, such as PIK3CD, FGFR3, TSC2, and RASGRP2, using real time PCR or conducting experiments in cell lines and mouse xenograft models. Together, the findings, published online today in Nature Communications, suggest that these splicing isoforms are associated with everything from tumor aggressiveness to targeted treatment resistance and poor patient outcomes.
"We found that the protein isoforms expressed in African Americans with prostate cancer do not always respond to targeted therapies, whereas these drugs were found to be effective in European Americans with prostate cancer and do end up killing off the cancer," senior author Norman Lee, a pharmacology and physiology researcher at George Washington University, said in a statement.
Although prostate cancer is a common cause of cancer death in American men in general, the team explained, there is evidence that the disease is not only diagnosed more often in African American men but also has a much higher mortality rate in this population, even after adjusting for socioeconomic contributors.
The researchers suspected that race-related alternative splicing might be a factor in such discrepancies, based on the prevalence of alternative splicing in human genes and their proposed role in cancer.
Using Affymetrix Human Exon GeneChip arrays, the team profiled gene splicing patterns in prostate cancer samples and matched normal prostate samples from the 35 African American or European American men.
More than 2,500 genes appeared to be differentially spliced between tumors from each population, including nearly 1,900 differential splicing events that were specific to the African American prostate tumors. Another 2,849 differential splicing events turned up in the normal prostate tissue from African Americans and European Americans.
When they focused in on genes with potential cancer contributions, based on gene ontology, prior associations, and other clues, the researchers found that the differentially spliced gene set was enriched for genes from pathways implicated in cancer formation — from growth factor to signalling pathways.
After applying real time PCR to a subset of the differentially spliced genes to validate them, the team took a closer look at a particular splice variant of PIK3CD that was over-represented in the African American samples, providing evidence that this version of the gene seemed to spur on invasiveness, proliferation, and small molecular inhibitor resistance in cell line experiments.
"The identification of novel splice variants as biomarkers and/or development of therapeutics targeting protein isoforms have the potential to reduce cancer disparities." Lee and his co-authors wrote.