Accurate preoperative staging remains a challenge in upper tract urothelial carcinoma (UTUC), with conventional imaging, ureteroscopic biopsy, and cytology providing limited ability to identify muscle-invasive disease before surgery.
A study published in European Urology Oncology evaluated whether integrating urine tumor DNA (utDNA) and plasma circulating tumor DNA (ctDNA) could improve molecular staging and risk classification in patients with nonmetastatic UTUC.
The study, titled “Integrated Urine and Plasma DNA Analysis for Molecular Staging of Upper Tract Urothelial Carcinoma,” was conducted by Maximilian Pallauf, Stephan Bronimann, Stanley Rapiey, Ezra Baraban, Shoujie Chai, Daniel S. Fischer, Vincent T. Bicocca, Trevor G. Levin, David J. McConkey, and Nirmish Singla.
Study Design and Molecular Testing
The investigators conducted a retrospective cohort study using a prospectively maintained database at a single tertiary referral center. The analysis included 41 patients with histologically confirmed, nonmetastatic UTUC who underwent radical nephroureterectomy or distal ureterectomy between 2015 and 2023 and had preoperative urine samples available.
Among the 41 patients, 21 (51%) had muscle-invasive disease on surgical pathology and 20 had non–muscle-invasive disease. Plasma samples were available for 22 patients, while matched formalin-fixed paraffin-embedded tumor tissue was available for 30. Overall, 56% of patients had received neoadjuvant chemotherapy, and 54% had a history of bladder cancer.
Urine and plasma samples were analyzed using the UroAmp next-generation sequencing assay. The primary analysis evaluated molecular staging against muscle-invasive disease confirmed by surgical pathology. Molecular staging was defined as either detection of a TP53 mutation in utDNA or a positive ctDNA result.
Urine DNA Reflected the Tumor Molecular Profile
Among 30 patients with matched urine and tumor tissue, utDNA mutations showed a mean overall concordance of 94% with mutations detected in the primary tumor when the same sequencing platform was used. Cross-platform comparison showed a mean concordance of 95%. Importantly, primary tumor mutations were identified in matched utDNA in all 30 tumors evaluated, supporting the potential use of urine as a noninvasive source of tumor genomic information.
Tumor-naive utDNA classification showed a sensitivity of 98% for detecting UTUC in the overall cohort and 95% among patients with muscle-invasive disease. In comparison, urinary cytology, available for 35 patients, had sensitivities of 40% overall and 33% among patients with muscle-invasive disease.
TP53 and ctDNA for Identifying Muscle-Invasive Disease
TP53 mutations were detected in utDNA in 50% of the cohort and were significantly enriched in patients with muscle-invasive disease. Detection of a TP53 mutation was associated with muscle-invasive UTUC with an odds ratio (OR) of 7.5 (95% CI, 2.0–33; p=0.004). Fifteen of 21 patients with muscle-invasive disease were positive for a TP53 mutation in utDNA, corresponding to a sensitivity of 71%.
By comparison, ureteroscopic biopsy had a sensitivity of 5% for identifying muscle-invasive disease, imaging had a sensitivity of 16%, and the combination of biopsy and imaging had a sensitivity of 19%. utDNA TP53 mutation testing was significantly more sensitive than standard diagnostic staging, while specificity was similar between the approaches.
Among the 22 patients with plasma available, utDNA-informed ctDNA testing was positive in four of 10 patients with muscle-invasive disease, corresponding to a sensitivity of 40%, and was negative in 11 of 12 patients with non–muscle-invasive disease, corresponding to a specificity of 92%.
Integrated Molecular Staging
When utDNA TP53 status and ctDNA status were combined as molecular staging, the approach was strongly associated with muscle-invasive disease on univariable analysis (OR 9.6; 95% CI, 2.5–44; p=0.002).
In a multivariable model incorporating molecular staging, conventional tumor stage, and tumor grade, molecular staging was the only statistically significant predictor of muscle-invasive disease (OR 8.5; 95% CI, 2.1–42; p=0.004). The investigators also compared molecular risk factors with standard-of-care risk factors when each was combined with diagnostic staging by biopsy and imaging.
Both approaches achieved a sensitivity of 81%. However, the molecular approach had a specificity of 70% compared with 25% for standard-of-care risk factors (p=0.008). Overall accuracy was 76% versus 54% (p=0.039), while the positive predictive value was 74% versus 53% (p=0.007). The negative predictive value was numerically higher with molecular risk assessment, at 78% compared with 56%, although this difference was not statistically significant (p=0.15).
Molecular Risk and Recurrence
Among patients who had not received neoadjuvant chemotherapy, six of 14 patients classified as molecularly high risk developed an extravesical recurrence, corresponding to a recurrence rate of 43%. No extravesical recurrences were observed among patients classified as molecularly low risk.
The authors suggested that integrating molecular testing with conventional assessment could help refine the identification of patients with higher-risk disease who may be considered for neoadjuvant treatment while potentially reducing overtreatment of patients with lower-risk tumors.
Limitations
The study included a relatively small cohort from a single institution, and matched plasma and tumor specimens were not available for every participant.
Patients with a history of bladder cancer were included, which could affect the specificity of urine-based testing because of field cancerization. The investigators addressed this by comparing genomic alterations in urine with those in resected tumor tissue and by performing subgroup analyses according to bladder cancer history and urine collection method.
The cohort also included patients both with and without neoadjuvant chemotherapy, introducing the possibility of pathological downstaging after treatment. The authors reported that analyses stratified according to neoadjuvant chemotherapy use did not show differences in test performance.
The study was partly supported by the Joseph and Betty Ann Migliara Fund for Urologic Cancer Research and the Paracelsus Medical University Research and Innovation Fund. Convergent Genomics provided in-kind sequencing and analysis services and reviewed the manuscript. The sponsor played a role in the design and conduct of the study, analysis of the data, and preparation and review of the manuscript.
Takeaway
In this 41-patient cohort, utDNA testing showed high sensitivity for nonmetastatic UTUC and closely reflected the somatic mutational profile of the primary tumor. Integrated utDNA and ctDNA testing provided a molecular framework for identifying muscle-invasive disease, with performance metrics that were equivalent or superior to standard-of-care evaluation. These findings support further investigation of urine- and plasma-based molecular testing for risk stratification and treatment planning in UTUC.
The full article is available in European Urology Oncology.
Find more videos on OncoDaily YouTube TV.
