Multi-gene profiling can lower need for prostate biopsies, but …
Posted Mar 07 2011 12:00am
One of the problems with the PSA test is that it identifies a possible risk for prostate cancer in many men each year who are subsequently shown by prostate biopsy to be at no risk for the actual clinical condition.
A new study from a research team at the Karolinska Institute in Sweden has explored whether risk profiles based on sets of multiple genetic markers can better differentiate between the men who really do need a biopsy and those who do not.
Aly et al. used data from the so-called Stockholm-1 cohort to carry out their study. The Stockholm-1 cohort is a group of 5,241 men, all of whom underwent a prostate biopsy between 2005 and 2007. For each of these men, PSA levels were recovered from their clinical records and questionnaires were used to obtain detailed family histories.
The research team then analyzed stored blood samples for a total of 35 single nucleotide polymorphisms (SNPs), representing changes in their genetic code, to develop a genetic risk score for each individual. The genetic risk scores were then used to develop a prostate cancer risk-prediction model.
Data from the genetic risk-prediction model were compared to prostate cancer risk data based on age, family history, total PSA, and %free PSA (a non-genetic model) , with the following results:
The genetic model suggested that 480 biopsies (22.7 percent) could be avoided.
Elimination of these 480 biopsies had a potential cost of a 3 percent increase of risk of non-diagnosis among men actually shown to have aggressive disease.
The genetic model was unable to distinguish between men with aggressive and non-aggressive forms of prostate cancer.
Despite all the continuing attempts to identify tests that will allow us to differentiate between the men who really do need treatment for their prostate cancer and those that very probably don’t (because at worst they have very indolent forms of the disease), we have yet to show clearly that multi-gene profiling can resolve this issue.
The “New” Prostate Cancer InfoLink suspects that it is more likely (at least in the immediate future ) that genetic profiling may be able to identify small subsets of men at high risk for aggressive forms of prostate cancer because of the presence of one or two specific genes (as suggested in the report last week about the SPINK1 gene ).