AUA 2018: Prospective Validation of a Diagnostic Urine Test For Bladder Cancer: The HEMAturia Study

San Francisco, CA ( Kim van Kessel, MD, presented a validation study of a urine-based diagnostic assay for bladder cancer: The HEMAturia study. The authors have previously published their findings on this test using a discovery cohort and retrospective validation cohort, reporting >90% sensitivity and 83-85% specificity.1,2 However, as those were both case enriched, they are moving to a prospective cohort. Now, they prospectively identified 1006 patients referred to Urology for microscopic or macroscopic hematuria in 6 hospitals in the Netherlands. The urine sample was collected prior to undergoing diagnostic cystoscopy.

The test identifies 3 mutations (FGFR3, TERT, HRAS) and 3 methylations (OTX1, ONECUT2, TWIST1) from urine samples with volumes as low as 2ml in some patients within their study cohort. 

Of 1092 patients who provided a voided urine specimen, only 1003 met inclusion criteria. Of these, 118 had urine specimens that could not be analyzed. This left 885 patients for analysis.

  • The 118 patients are an issue - ~10% of patients whose urine didn’t yield results
  • The presenter herself addressed this – they did not require a specific volume of urine be submitted. While volume isn’t necessarily the issue, it is the concentration that matters. Despite that, more volume increased yield. So, in the future, they will specify volume to be submitted and request 2 samples (has been shown to increase yield).
55% of patients presented with macroscopic hematuria, and 45% presented with microscopic hematuria. 59% were male. 115 patients were identified to have urothelial cancer. Patients with cancer were more likely to have gross hematuria. Majority were bladder cancers (96%). Most were Ta or T1. 42% were grade 3. 13% had concomitant CIS.

The initial model combined age + methylation + mutation markers. The assay resulted in an AUC of 0.92, with 88% sensitivity and 80% specificity. Negative predictive value was 98%. 

However, when they substituted nature of hematuria (macro or microscopic) instead of age, the assay resulted in an AUC of 0.95, with 93% sensitivity and 81% specificity. Negative predictive value was 99%.

In terms of missed diagnoses, there were 6/101 in patients with macroscopic hematuria and 2/14 in patients with microscopic hematuria. There were 9 false positives in the entire cohort (all in microscopic hematuria). All upper tract tumors were detected. 

These results are truly remarkable, and in some ways, almost too hard to believe! Yet, as they have demonstrated in a large prospective dataset the efficacy of this test, it is very convincing as the real-deal. The assay will now be developed for clinical use, and is being developed commercially by MDxHealth, Inc.

The development of such an assay has long been a dream for clinicians who see large volumes of hematuria patients. The current standard of care, which includes axial imaging, cystoscopy, and urine cytology for some patients, is costly, time-consuming, and not 100% diagnostic. Having a urine-based biomarker test to select which patients should undergo more intensive testing is a potential game-changer for our field. In this study alone, nearly 88% of patients might have been spared cystoscopic evaluation if the biomarker test were used as an initial screening tool.

Of course, the test still needs to be proven in the screening role, but it is likely that it will play an important role in future hematuria algorithms. Surely, this test will also be studied for use in patients under bladder cancer surveillance and potentially to even evaluate patients after neoadjuvant chemotherapy to assess the need for radical cystectomy (identifying T0 patients).

Limitations / Discussion Points:

1. One of the audience members asked about logistics:
   - volume of urine received ranged from 2-100 mL
   - concentration is more important than volume
   - 2 samples better than 1
   - Results take about 3 days to return

1. van Kessel KE, Beukers W, Lurkin I, Ziel-van der Made A, van der Keur KA, Boormans JL, Dyrskjøt L, Márquez M, Ørntoft TF, Real FX, Segersten U, Malats N, Malmström PU, Van Criekinge W, Zwarthoff EC. Validation of a DNA Methylation-Mutation Urine Assay to Select Patients with Hematuria for Cystoscopy. J Urol. 2017 Mar;197(3 Pt 1):590-595. doi: 10.1016/j.juro.2016.09.118. Epub 2016 Oct 13
2. van Kessel KE, Van Neste L, Lurkin I, Zwarthoff EC, Van Criekinge W. Evaluation of an Epigenetic Profile for the Detection of Bladder Cancer in Patients with Hematuria. J Urol. 2016 Mar;195(3):601-7. doi: 10.1016/j.juro.2015.08.085. Epub 2015 Aug 29.

Presented by: Kim van Kessel, MD, MSc, Erasmus Medical Center
Co-Authors: Joep de Jong, Angelique Ziel-van der Made, Rotterdam, Netherlands, Hossain Roshani, The Hague, Netherlands, Stefan Haensel, Josien van Montfrans-Wolterbeek, Egbert Boevé, Rotterdam, Netherlands, Eric Oomens, Breda, Netherlands, Niels van Casteren, Capelle a/d IJssel, Netherlands, Joost Boormans, Rotterdam, Netherlands, Wim Van Criekinge, Irvine, CA, Ellen Zwarthoff, Rotterdam, Netherlands

Written by: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto, @tchandra_uromd at the 2018 AUA Annual Meeting - May 18 - 21, 2018 – San Francisco, CA USA
email news signup