This open-label, randomised, multicentre, phase 2 study (OHAI-NMIBC-01) compares toripalimab plus sequential intravesical gemcitabine followed by mitomycin C (GEM→MMC) with toripalimab alone in adults with BCG-unresponsive or BCG-intolerant high-risk non-muscle-invasive bladder cancer (HR-NMIBC). Two prespecified cohorts are analysed: (1) CIS cohort (CIS with/without Ta/T1) and (2) non-CIS cohort (high-risk Ta/T1 without CIS). In the combination arm, intravesical GEM→MMC is given weekly for 6 weeks (induction) and, for patients without recurrence at the first tumour assessment (~month 3), monthly maintenance continues up to 24 months or until progression/unacceptable toxicity; toripalimab IV every 3 weeks starts during the first intravesical cycle and continues up to 24 months or until progression/unacceptable toxicity. The monotherapy arm receives toripalimab IV every 3 weeks up to 24 months or until progression/unacceptable toxicity. Cystoscopy and urine cytology are performed every 3 months; imaging every 24 weeks. Primary endpoints are 3-month complete response (CR) rate in the CIS cohort and median recurrence-free survival (RFS) in the non-CIS cohort. Secondary endpoints include landmark CR, PFS and OS, RFS/HG-RFS landmarks in the non-CIS cohort, and safety (CTCAE v5.0). Exploratory analyses will assess outcomes by protocol-defined PD-L1 status. Approximately 106 participants will be enrolled at multiple sites in China.

Rationale and Objectives. A substantial proportion of HR-NMIBC patients are BCG-unresponsive or BCG-intolerant and face early high-grade recurrence and consideration of cystectomy. This trial evaluates whether adding sequential intravesical chemotherapy (gemcitabine followed by mitomycin C in the same visit) to systemic PD-1 blockade (toripalimab) improves outcomes versus toripalimab alone.

Design. Prospective, open-label, randomised (1:1), parallel-group, phase 2 study at multiple centres in China. Treatment continues until recurrence/progression, unacceptable toxicity, withdrawal, or completion of 24 months.

Interventions.

Arm A (Combination): Toripalimab 240 mg IV every 3 weeks (Q3W) up to 24 months plus intravesical GEM→MMC: gemcitabine 1,000 mg retained ~60 min and drained, then mitomycin C 40 mg retained ~60 min; administered weekly for 6 weeks (induction) and then maintenance every 4 weeks (Q4W) up to month 24 if no high-grade recurrence.

Arm B (Monotherapy): Toripalimab 240 mg IV Q3W up to 24 months.

Assessments. Cystoscopy and urine cytology at months 3, 6, 9, 12 and every 3 months thereafter; abdominopelvic/upper-tract CT or MRI every 24 weeks; routine laboratory tests prior to dosing/instillation. (Exploratory tissue/urine/blood sampling may be performed per protocol.)

Primary Endpoints.

CIS cohort: CR rate at month 3 (proportion). CR is met by any of the following protocol-specified scenarios indicating no high-grade bladder disease: (a) negative urine cytology and negative cystoscopy; (b) negative cytology with cystoscopic lesions that are benign or low-grade Ta on biopsy; or (c) negative cystoscopy with positive cytology attributed to tumour in the upper tract or prostatic urethra and random bladder biopsies negative.

Non-CIS cohort: Median RFS, defined as time from start of treatment to high-grade Ta recurrence, any T1, or new CIS. Recurrence requires cystoscopic suspicion confirmed by histopathology; if cytology becomes positive prior to histologic confirmation, the recurrence date is set at the first positive cytology once recurrence is subsequently confirmed.

Secondary Endpoints.

CIS cohort: CR rates at months 6, 12, 18, and 24; PFS rates at the same landmarks (progression defined as lamina propria invasion from Ta/CIS to T1, progression to ≥T2, or new nodal/distant metastasis); OS at month 24, EOT+6 months, and EOT+12 months; incidence of adverse events (AEs).

Non-CIS cohort: RFS rates at months 6, 12, 18, and 24; high-grade RFS (HG-RFS) rates at the same landmarks (high-grade recurrence defined as Tis or Ta/T1 high-grade, or muscle-invasive disease at TURBT or cystectomy); PFS rates at months 6/12/18/24; OS at month 24, EOT+6 months, and EOT+12 months; AE incidence.

Exploratory Endpoints.

CIS cohort: 3-month CR rate in PD-L1-positive and PD-L1-negative subgroups (per protocol-specified assay and criteria; no prespecified CPS threshold).

Non-CIS cohort: Median RFS in PD-L1-positive and PD-L1-negative subgroups (recurrence definitions as above).

Sample Size and Oversight. Approximately 106 participants will be randomised 1:1. Safety will be monitored throughout; a Data Monitoring Committee oversees participant protection and study conduct.

Inclusion Criteria:

1. Age ≥18 years; sex: all; signed written informed consent by the participant or legally authorised representative.

2. Histologically confirmed high-risk non-muscle-invasive bladder cancer (HR-NMIBC), defined as any T1, high-grade Ta, and/or carcinoma in situ (CIS).

3. BCG-intolerant (unable to continue BCG because of severe adverse reactions) or meeting at least one definition of BCG treatment failure:

   1. Persistent or recurrent CIS within 12 months after completion of adequate BCG (with or without concomitant NMIBC);
   2. Recurrent high-grade Ta/T1 within 6 months after completion of adequate BCG;
   3. High-grade T1 at the first evaluation after BCG induction (~3 months);
   4. Ta high-grade and/or CIS present or recurrent at ~3 months after receiving ≥5 BCG instillations.

   Adequate BCG, for the purposes of this protocol, is defined as receipt of at least 5 of 6 induction instillations (maintenance not required).

4. ECOG performance status 0-2.

5. Adequate organ function per protocol laboratory criteria.

6. No intravesical chemotherapy or immunotherapy between the most recent cystoscopy/TURBT and study start; a single immediate postoperative intravesical chemotherapy at the time of the most recent cystoscopy/TURBT is allowed during screening per local practice.

7. Willing and able to comply with study procedures.

Exclusion Criteria:

1. Muscle-invasive bladder cancer (T2-T4).
2. Low-grade (LG) recurrence during or after BCG therapy.
3. Concomitant upper tract urothelial carcinoma, or lymph-node/distant metastasis.
4. Indwelling ureteral stent or known vesicoureteral reflux.
5. Contraindications to intravesical instillation, including within 2 weeks after TURBT, bladder perforation, symptomatic urinary tract infection, or gross haematuria.
6. Known hypersensitivity or contraindication to gemcitabine, mitomycin C, or toripalimab.
7. Systemic chemotherapy, small-molecule targeted therapy, or radiotherapy within 2 weeks before first study treatment.
8. Prior immune checkpoint inhibitor therapy.
9. Pregnant, planning pregnancy, or breastfeeding women.
10. Ongoing acute or chronic systemic infection, or history of active tuberculosis.
11. Other malignancy requiring active treatment.
12. Any condition that, in the investigator's judgment, makes participation not in the patient's best interest or could confound study results.

Study Population Adults with BCG-unresponsive or BCG-intolerant HR-NMIBC treated at participating centres in China.

The sponsor and investigators intend to share de-identified IPD but the final plan (scope, timing, and access mechanism) is pending institutional approvals. This record will be updated once the IPD sharing policy is finalised.