In a simple urine test designed by scientists at Stanford Medicine and the VA Palo Alto Health Care System, a revolution in how doctors detect and treat bladder cancer is taking shape. Unlike other tests, which rely solely on DNA mutations from the cancer, the approach to uRARE-seq looks for fragments of RNA from tumor cells that have been shed into the urine, which brings a more immediate read on what the cancer is up to and how it might behave.
The new test was 95 percent accurate at identifying people with localized bladder cancer and98 percent accurate at excluding the disease when tested on 683 urine samples from patients and healthy volunteers. These results were better than those achieved with the conventional method of urine cytologythe microscopic examination of shed tumor cellsand other urine tests, which detect DNA. Still, the new assay could do more than just identifying this type of cancer; it could differentiate between low-grade and high-grade disease and detect evidence of residual tumor following radical cystectomy or intravesical BCG immunotherapy (the most common upfront management that involves injecting a weakened form of tubercle bacillus into the bladder to activate an immune response).
Most impressive though, is the ability of the assay to predict response before treatment, before BCG even being administrated. Upon examining the results, it was observed that responding patients had RNA signals in their urine strongly enriched for genes connected to T-cell activity and immune signaling pathways. Non-responders showed stronger urine signals for rapidly dividing tumor cells. On this basis, the group designed a biomarker that accurately predicted the likelihood of a tumor to respond to either immunotherapy or chemotherapy(used after failure or where BCG is unavailable) using pretreatment samples from 114 patients. The area under the receiver operator curve for the model was 0.93.
The test opens the door to potentially eliminate all those repeated, invasive camera tests, known as cystoscopies, which patients associate with the current tools used for diagnosis and monitoring. A cystoscopy is uncomfortable and stressful for patients and packs a small risk of infection or injury. An accurate non-invasive tool would make follow-up testing simpler and perhaps more sensitive to actual tumour activity. The RNA method also seems less reliant on certain potentially misleading mutation signals related to field effects.
Bladder cancer is one of the more common cancers in the US and many are diagnosed early, still within the lining of the bladder. Yet, there is a high rate of recurrence and determining who takes more aggressive treatment postoperatively can be a challenge. Molecular indicators of remaining disease and likely responding therapy could avoid unnecessary side effects in some and give others a head start.
Results were published in Nature Medicine in early October 2026. The authors state that more large prospective studies are required to determine whether this tests can be implemented into clinical practice.