Reply to the impact of biomarkers in multivariate algorithms for bladder cancer diagnosis in patients with hematuria


Reply to the Impact of Biomarkers in Multivariate Algorithms for Bladder Cancer Diagnosis in Patients with Hematuria

We were very interested to read the letter by Drs. Cai, Malossini, and Bartoletti. This group described, in a rigorously conducted study, how microsatellite analyses (MSA) of urine sediments had sufficient sensitivity and specificity to be considered as a screening and monitoring assay for non–muscle-invasive urothelial carcinoma.1 They point out that MSA correctly classified 78.5% of benign prostatic hyperplasia (BPH) patients as controls. On this basis, they suggest that the addition of MSA to the multivariate diagnostic bladder cancer algorithm we described2 could decrease the number of false-positive results attributable to patients with BPH. In response to their first question, the unpredictable nature of the multiple direct and indirect interactions in carcinogenic pathways make it impossible to intuitively predict3 to what extent the addition of MSA or CD14, a novel biomarker for BPH,4 would enhance the sensitivity or specificity of the multivariate algorithm we described. A further consideration is that combining diagnostic assays is cumbersome and would detract from our ultimate goal, namely to create a high-throughput assay to stratify risk in patients who present with hematuria in the primary care setting. We should point out that the study we published in Cancer2 is the first step in the process to create the high-throughput assay. This study demonstrated that a multivariate algorithm significantly improved on the diagnostic accuracy provided by demographics.2 We have planned a larger study that will recruit approximately 1000 patients: 666 controls and 333 patients with urothelial cancer. The goal of this future study will be to create a biochip arrayed with the proteins that, in combination, achieve the highest specificity and sensitivity to identify urothelial cancer in patients presenting with hematuria. We have designed both studies to include the full range of pathologies, inclusive of patients with both non–muscle-invasive and muscle-invasive urothelial cancer, because this reflects the population in which the biochip would ultimately be applied. It is interesting to note that Bartoletti et al1 touch on another important aspect to distinguishing early cancers: inflammation. Inflammation plays a role in bladder and perhaps also in prostate carcinogenesis.5 Our understanding of the role of inflammation in carcinogenesis is elementary and it is as yet unknown at what point “benign” confounding pathologies such as BPH switch to become early cancers. Perhaps it is wise to be equivocal as to what constitutes a low-risk “benign” pathology when considering BPH or any other inflammatory pathologies that present with hematuria.


Drs. Ruddock and Williamson are named inventors on British Patent No. 0916193.6, which protects the biomarkers in the algorithms. In addition, Dr. Ruddock is an employee of Randox Laboratories Ltd, who undertook the biomarker analyses using multianalyte biochip technology.