MedicalResearch.com Interview with:
Dr Helen Ross-Adams
Cancer Research UK, London
Medical Research: What is the background for this study? What are the main findings?
Dr. Ross-Adams: Prostate cancer is the most common non-skin cancer in men in both the UK and US. At the moment, prostate cancer is diagnosed and monitored mainly on the basis of blood tests for prostate specific antigen (PSA), a protein in the blood. MRI scans and examination of biopsy tissue samples under a microscope are also used to decide on the best course of action for each patient.
Despite all this, as a community, we still struggle to reliably predict which men with an initial diagnosis of prostate cancer will go on to have a fast-growing, aggressive form of the disease (a ‘tiger’) from men who will have a much slower-growing form of the disease that won’t really cause problems in the man’s lifetime (a ‘pussycat’). This means some men may get treatment they don’t need, while others could benefit from earlier, more intensive treatment.
With this in mind, we studied a total of 250 men with prostate cancer and tested their tumour and healthy tissues at the molecular level. The idea was two-fold:
- Could we identify different sub-types of prostate cancer using this genetic information, and
- Could we link any of the sub-types we did find with other patient characteristics that clinicians would normally have, like histological staging information or PSA test results?
We looked at their DNA, to see whether any regions were deleted or repeated (copy number alterations), and we also measured the activity levels of thousands of genes in the tumour and healthy prostate tissues (gene expression). Each of these approaches on their own can be used to stratify patients, but we decided to combine this information and hopefully find genes that had a big impact on prostate cancer.
Using this approach, we identified five different subtypes of prostate cancer, each with their own ‘molecular profile’:
- One group had lots of DNA deletions and only low levels of certain genes
- Another had lots of repeated DNA with high levels of associated genes
- Two more groups had very ‘quiet’ genomes, with very few changes at the DNA level, and not much disruption at the gene expression level
- The fifth and final group had an intermediate amount of copy number changes (DNA level), but no major changes at the gene expression level (mRNA level)
When we correlated these different molecular subtypes with the patients’ standard post-surgery follow-up data (the results of 6-monthly PSA tests), we found that these subtypes predicted how well a patient would do after surgery. We ultimately identified 100 key genes (a gene signature) that were most useful in classifying men into one of the 5 cancer subtypes we identified.
This was derived from 150 men in Cambridge, UK. To check our findings, we repeated the same work in a group of 100 men from Stockholm, Sweden who had also had prostate surgery, and found that the 100 gene signature worked just as well – it subdivided the men into 5 different groups, each with different rates of relapse. In both cases, men with the most genetic alterations had the greatest chance of relapsing after surgery.
MedicalResearch.com Interview with:
Dr. David Brent MD
Department of Psychiatry
Western Psychiatric Institute and Clinic
University of Pittsburgh Medical Center
Pittsburgh, Pennsylvania
Medical Research: What is the background for this study?
Dr. Brent: Youth with a parent with a history of depression are at increased risk for having a depressive episode themselves.
Medical Research: What are the main findings?
Dr. Brent: Those who received a cognitive behavioral educational group program were less likely to have had a depressive episode, and were functioning better than those who did to receive the program 6 years later, especially if their parent was NOT depressed at the time that they received the program. If the parent was depressed then the program was no better than usual care.
