Quick answer
Fudan University Shanghai Cancer Center has developed a urine-based test that measures RNA and uses an algorithm to detect urothelial carcinoma — cancer of the lining of the bladder and urinary tract. The hospital reports overall diagnostic sensitivity of 91.37% and specificity of 95.09%, and says the system has completed medical device registration and is in clinical use [1][2]. The gap it is trying to close: standard white-light cystoscopy finds only about 70% of flat lesions such as carcinoma in situ, and is described as missing around 70% of carcinoma-in-situ recurrences [2]. The test is reported as the first of its kind [4].
Key takeaways
- The test uses a urine sample — no instrument is passed into the bladder — and combines RNA measurement with an algorithm [1].
- Reported performance: 91.37% sensitivity and 95.09% specificity, described as internationally leading across stages of urothelial carcinoma [1][2].
- The comparison the hospital makes is with white-light cystoscopy, which it puts at roughly 70% sensitivity and says misses about 70% of carcinoma-in-situ recurrences [2].
- It has completed medical device registration and is in clinical use at the centre [1][2].
- The hospital reports five-year survival of 74% for the bladder cancer patients it treats, alongside 82.6% for prostate and 77.1% for kidney [2].
- These are the hospital's own figures. We did not find a peer-reviewed publication of the test's performance [1][2].
The condition in context: bladder cancer and the problem of watching for it
Most bladder cancer starts in the urothelium, the lining of the bladder. The usual first sign is blood in the urine, which is why many cases are found early — but finding them is only half the job. Bladder cancer comes back often, so patients need surveillance for years, and that surveillance has traditionally meant repeated cystoscopy: a scope passed through the urethra into the bladder, uncomfortable enough that people skip appointments.
| Region | What the data shows | Source |
|---|---|---|
| Worldwide | 614,298 new bladder cancer cases in 2022 and 220,596 deaths. | World Cancer Research Fund, from GLOBOCAN 2022 [3] |
| China | 92,883 new cases in 2022 — more than any other country — and 41,367 deaths, also the highest. Age-standardised rate 3.4 per 100,000. | WCRF / GLOBOCAN 2022 [3] |
| United States | 80,404 new cases in 2022 at an age-standardised rate of 10.5 per 100,000. The American Cancer Society's estimates for 2026 are about 84,530 new cases and 17,870 deaths. | WCRF / GLOBOCAN 2022 [3]; ACS Cancer Facts & Figures 2026 [5] |
| Japan | 34,568 new cases in 2022 (age-standardised rate 7.0 per 100,000) and 10,928 deaths. | WCRF / GLOBOCAN 2022 [3] |
China has the highest number of cases but a low age-standardised rate, while the United States has fewer cases at a rate roughly three times higher — a reminder that case counts and rates measure different things and should not be compared directly [3].
Against that background, a urine test with these accuracy figures would matter in two places: initial assessment of people with blood in their urine, and the long follow-up that bladder cancer survivors go through [1][2].
What was reported
The work was presented around the 2026 Pujiang Urological Oncology Conference by Fudan University Shanghai Cancer Center. The team was led by Ye Dingwei, director of the centre's urological oncology centre [1][2].
The hospital describes the system as combining RNA measurement from urine with an artificial-intelligence algorithm, and reports that it performs consistently across different stages of urothelial carcinoma. It says the system has completed medical device registration and has entered clinical use [1][2].
The same report describes related work: a Nectin-4 targeted peptide probe developed with the nuclear medicine department to assess target expression without a biopsy, intended to identify patients suitable for antibody-drug conjugate therapy; and organ-preservation strategies including a PD-1 plus BCG approach for high-risk non-muscle-invasive bladder cancer [2].
Sensitivity is the share of people who have the disease and are correctly identified: at 91.37%, roughly 9 in 100 cancers would still be missed. Specificity is the share of people without the disease who are correctly cleared: at 95.09%, about 5 in 100 healthy people would get a false alarm. So a positive result still needs confirming, and a negative result does not fully rule cancer out [1][2].
Why it matters for people considering care in China
| What it tells you | What it does not tell you |
|---|---|
| A urine-only option exists at one Shanghai centre for urothelial cancer assessment, and is in clinical use [1][2]. | Whether it has been validated independently or in a peer-reviewed study — we did not find one. |
| The stated comparison — white-light cystoscopy at about 70% sensitivity — is a real clinical problem worth solving [2]. | Whether this test would replace a cystoscopy in your situation, or only supplement it. |
| The centre reports strong survival figures for the patients it treats [2]. | How much of that is the test, how much is surgical and systemic treatment, and how much is case mix. |
Availability and limits
- The figures are the hospital's own. We found no peer-reviewed publication of the test's sensitivity and specificity, and no independent validation.
- Registration is Chinese. The reported medical device registration covers China. We found no information about approval or availability elsewhere.
- Not a population screening test. On the reporting so far, this is for people already under assessment or surveillance, not a general check for people with no symptoms.
- A positive still needs a scope. At 95.09% specificity, false positives occur, and any positive result would still require confirmation — usually by cystoscopy and biopsy.
- Cost and access for international patients are not published. None of the sources gives a price or says whether non-residents can be tested. Ask the hospital directly.
- Reported survival figures are centre-specific. 74% five-year survival for bladder cancer refers to patients treated at this hospital, not to bladder cancer generally [2].
What to ask before booking anything
| Question | Why it matters |
|---|---|
| Is the urine RNA test appropriate for my situation — initial diagnosis, or surveillance after treatment? | The reporting covers assessment and screening, not every scenario. |
| If the result is positive, what happens next? | You need to know whether a cystoscopy and biopsy will still be required. |
| How much does the test cost, and is it covered? | New molecular tests are often self-funded. |
| Can the sample be sent from abroad, or do I need to attend in person? | Determines whether this is worth travelling for. |
| Is English-language reporting available? | Results you cannot read are of limited use to your doctor at home. |
Questions patients ask
What is urothelial carcinoma?
It is cancer of the urothelium — the lining of the bladder, ureters and renal pelvis. Bladder cancer is the most common form, which is why the two terms are often used together [1][2].
How is bladder cancer usually checked for?
Typically with cystoscopy, a thin camera passed through the urethra into the bladder, often with imaging and urine cytology. The hospital notes that standard white-light cystoscopy has limited sensitivity — about 70% — for flat lesions such as carcinoma in situ, and can miss around 70% of carcinoma-in-situ recurrences [2].
Does a negative urine RNA test rule out cancer?
No. At 91.37% sensitivity, roughly nine in a hundred cancers would still be missed. It should be used alongside, not instead of, the judgement of a urologist [1].
Is this test available outside China?
We found no published information on approval or availability outside China, where it is reported to have completed medical device registration [1][2].
Can I use it as a general health check?
Nothing in the reporting supports that. It has been described for screening and assessment of urothelial carcinoma in people who are already under urological care, not as a screening test for people with no symptoms [1][2].