Drug-conjugate plus immunotherapy clears high-risk bladder tumours in small trial

A combination of disitamab vedotin, an antibody-drug conjugate aimed at the HER2 protein, and the immune-checkpoint drug tislelizumab cleared high-risk disease in most patients in a small, single-arm trial of bladder cancer that could not be surgically removed through the urethra [s1]. The result is early — 28 patients, no comparison group — and the authors frame it as support for a randomised trial rather than as proof of benefit [s1].

The patients had very-high-risk non-muscle-invasive bladder cancer, a category that includes extensive high-grade T1 tumours, disease with carcinoma in situ, unfavourable tissue types, or cancer unresponsive to bacille Calmette-Guerin (BCG) therapy [s1]. When such tumours cannot be resected through the urethra, the standard answer is removal of the bladder; patients who are ineligible for that surgery or decline it have had no established drug alternative [s1]. Disitamab vedotin links an antibody that homes to the HER2 protein to a cell-killing payload, so that the chemotherapy is delivered preferentially to tumour cells carrying the target; tislelizumab releases a brake on the immune system by blocking the PD-1 protein [s1]. Pairing the two tests whether a targeted chemotherapy and an immune drug can control the disease without surgery [s1].

What the trial did

This was an open-label, single-arm, phase 2 study with a Simon two-stage design, run at one academic hospital from August 2022 to May 2024, with a median follow-up of 29.4 months [s1]. It enrolled adults whose HER2-positive (ERBB2-positive) tumours met 2019 European Association of Urology very-high-risk criteria and could not be resected transurethrally, and who were ineligible for or declined bladder removal [s1]. Patients received intravenous disitamab vedotin (120 mg on day 1) plus tislelizumab (200 mg on day 2) every three weeks for three cycles; those who responded continued for up to eight disitamab vedotin cycles and 17 tislelizumab cycles [s1][s2]. The primary outcome was a complete response of high-risk disease, meaning no high-grade tumour, T1 disease, carcinoma in situ or progression [s1].

What it found

Of 137 patients screened, 28 were enrolled — median age 72 years (interquartile range 67 to 74), and 25 (89%) were male [s1]. Two patients withdrew before assessment and were counted as non-responders in the all-treated analysis, leaving 26 who were efficacy-evaluable [s1].

A complete response of high-risk disease was achieved in 19 of 26 evaluable patients (73.1%; 95% confidence interval 54.8% to 91.3%), meeting the trial's prespecified 70% target [s1]. Counting all 28 treated patients, 19 responded (67.9%; 95% confidence interval 49.4% to 86.3%), which exceeded the trial's threshold of 17 responses [s1]. Both analyses met the primary criterion [s1]. Among responders, 83.5% still had a sustained response at 24 months (95% confidence interval 68.0% to 100.0%), and no patient progressed to muscle-invasive or metastatic disease [s1].

Treatment-related adverse events occurred in 25 patients (89.3%), most often paraesthesia (12, 42.9%), hair loss (11, 39.3%) and itch (10, 35.7%) [s1]. Four patients (14.3%) had grade 3 to 4 events, two (7.1%) stopped treatment, and none had a grade 5 event [s1].

How to read it

The design sets the ceiling on what this can show. A single-arm study with no comparison group cannot establish that the combination is better than bladder removal, than another drug regimen, or than close surveillance [s1]. With 28 patients from one hospital, the confidence intervals are wide — the all-treated response interval runs from roughly 49% to 86% — and the findings may not transfer to other settings or to HER2-negative disease, which was not studied [s1]. The trial also enrolled only tumours that were HER2-positive and transurethrally unresectable, a specific slice of bladder cancer [s1].

Within those limits, the signal is consistent: a high complete-response rate that met both prespecified statistical bars, responses that largely held at two years, no progression to invasive or metastatic disease among those followed, and side-effects that were common but mostly low-grade [s1]. The two preassessment withdrawals were deliberately counted as non-responders, a conservative choice that works against the drug rather than flattering it, and even so the all-treated analysis cleared its threshold [s1]. The authors present it as a basis for randomised testing, not as a new standard of care [s1].

What to watch

The open questions are whether a randomised trial confirms the benefit against established options, how durable the responses are beyond two years, and whether the approach helps patients whose tumours are not HER2-positive [s1][s2]. This article describes research and is not medical advice; decisions about bladder-cancer treatment belong with treating clinicians.