ORLANDO, FL USA (UroToday.com) - Dr. Dan Theodorescu from the University of Colorado gave a “call to arms” for urologists to be aware of and develop biomarkers for bladder cancer.
He began by contrasting the past, present, and future of risk stratification. In the past, empirical stratification was based on history and physical examination of the patient, whereas currently “stratified medicine” is used, which compares individual patients to groups of patients with similar clinical phenotype. In the future, he said that we will use biomarkers to administer personalized medicine to bladder cancer patients, stating that, “biomarkers are the pillars of personalized medicine.” Tissue, blood, and urine can be used to measure mutations, DNA hypermethylation, SNPs, gene expression, copy number variation, and protein expression or modification both singly and in multiplex fashion. These markers can be used to create predictive (“who to treat”) and prognostic (“what to give”) factors to guide clinical management.
He provided a few examples of markers in development. Three studies identified STAG2 mutations which alter chromatid cohesion and segregation during mitosis. It is mutated in up to 30% of bladder cancer specimens. Ironically, when mutated, it was associated with a poor prognosis, however, low levels of STAG2 protein expression were associated with a good prognosis. He harkened back to early p53 studies in bladder cancer patients that stratified patients into good and poor prognosis but were later invalidated in prospective studies. He provided another cautionary note when he highlighted a recent publication showing that randomly picked markers from a large panel were just as predictive as markers that were picked from a statistically designed set. He urged prospective validation of marker sets, and highlighted what he considered was a properly done study. This study was used to predict node-positive patients, at the time of cystectomy, based on markers at the time of transurethral resection of bladder tumor (TURBT). The study had a training set to develop the marker panel which was validated in multiple independent cohorts, and which reproducibly showed statistical significance.
He closed by discussing challenges in bladder marker design, including prospective evaluation of marker sets, standardizing metrics to measure markers, and incorporation of biomarkers into multivariate nomograms.
Presented by Dan Theodorescu, MD, PhD at the American Urological Association (AUA) Annual Meeting - May 16 - 21, 2014 - Orlando, Florida USA
University of Colorado, Denver USA
Written by Philip Abbosh, MD, PhD, medical writer for UroToday.com