Data from trials like Niagara suggests a powerful new paradigm for assessing treatment success. Combining urine tumor DNA (uTDNA) for local disease and circulating tumor DNA (ctDNA) for systemic relapse offers a more dynamic view than traditional pathology and is poised to become the superior surrogate endpoint in bladder cancer trials.

Circulating tumor DNA (ctDNA) is a powerful biomarker for identifying high-risk bladder cancer patients. However, its imperfection presents a new clinical dilemma: with a ~12% relapse rate even in ctDNA-negative patients, clinicians must decide whether to withhold adjuvant therapy and accept that risk, or overtreat the 88% who are likely cured.

Personalized, tumor-informed ctDNA approaches (like Signatera) are more specific and sensitive for bladder cancer than general panel-based assays. This is because bladder cancer has a distinct mutational landscape and because panel approaches can yield false positives from clonal hematopoiesis, making the personalized technique essential for accurate results in this disease.

The original Signatera assay used 16 personalized probes based on whole-exome sequencing to find ctDNA. The next-generation version, based on whole-genome sequencing, expands this to 64 probes. This is expected to significantly increase sensitivity, detect molecular relapse earlier, and provide a longer window for clinical intervention.

While circulating tumor DNA (ctDNA) is a powerful prognostic marker, it is not yet part of the formal "clinical complete response" definition for bladder-sparing trials. Experts lack data on its ability to predict the superficial, non-muscle invasive relapses common in this setting.

Beyond a simple positive/negative result, the quantitative level of ctDNA is highly prognostic in bladder cancer. Similar to PSA in prostate cancer, higher ctDNA levels correlate with a significantly worse prognosis, offering a more nuanced risk assessment tool than a binary test.

The TRACK-ER study reveals a critical weakness of tumor-informed ctDNA monitoring: a 16% failure rate. This occurs when there's insufficient tumor tissue or too few personalized variants to track. This technical barrier poses a significant obstacle to widespread clinical implementation, highlighting the need for more robust or alternative assay technologies for all patients to benefit.

Tumor-informed assays like Signatera sequence a patient's tumor to create a personalized test, making it highly sensitive but taking 3-4 weeks. Tumor-uninformed assays are faster (1 week) but less sensitive as they screen for a generic panel of cancer mutations.

Tumor-informed ctDNA assays, which require a tissue sample, are highly sensitive and well-suited for the adjuvant setting where tissue is available and time is less critical. In the metastatic setting, logistical challenges and the need for faster results make this approach less practical.