Neoadjuvant cisplatin-based chemotherapy followed by radical cystectomy with lymph node dissection remains an important treatment approach for suitable patients with operable muscle-invasive bladder cancer. More recently, perioperative chemoimmunotherapy has improved outcomes in this setting, but pathologic complete response rates have remained below 45% in prior trials.
Intravesical Bacillus Calmette-Guérin is widely used in high-risk non–muscle-invasive bladder cancer. However, its role in muscle-invasive disease has been limited, partly because of modest historical experience and concerns about systemic dissemination during chemotherapy-induced immunosuppression.
SAKK 06/19 evaluated whether adding intravesical recombinant Bacillus Calmette-Guérin, known as rBCG or VPM1002BC, to perioperative chemoimmunotherapy could improve pathologic response in patients with operable muscle-invasive bladder cancer.
The original article, titled “Addition of Intravesical Recombinant Bacillus Calmette-Guérin to Perioperative Chemoimmunotherapy in Muscle-Invasive Bladder Cancer: Primary Analysis of the Single-Arm Phase II Trial SAKK 06/19,” was published in the Journal of Clinical Oncology on May 29, 2026.
Authors: Richard Cathomas, Ulf Petrausch, Stefanie Hayoz, Martin Spahn, Lukas Bubendorf, Karin Schäuble, Räto T. Strebel, Philipp Niederberger, Frank Stenner, Anita Hirschi, Ursula Vogl, Anja Lorch, Petros Tsantoulis, Stefanie Aeppli, Andreas Erdmann, Angela Fischer Maranta, Christian Fankhauser, Andreas M. Hötker, Antonia M. Pausch, Petra Herzig, Victor Le Gall, Alfred Zippelius, Sabrina Chiquet, Sacha I. Rothschild, and Cyrill A. Rentsch.
About SAKK 06/19
SAKK 06/19 was a multicenter, investigator-initiated, open-label, single-arm phase 2 trial conducted at 10 sites within the Swiss Cancer Institute network in Switzerland.
The trial enrolled patients with histologically confirmed urothelial carcinoma of the bladder, clinical stage cT2-T4a N0-1, without distant metastases. Eligible patients had WHO performance status 0 or 1, were considered suitable for curative multimodality treatment including radical cystectomy with lymph node dissection, and were eligible for cisplatin-based chemotherapy.
The treatment plan included intravesical rBCG on days 1, 8, and 15. Atezolizumab was started on day 1 and given every 3 weeks for four doses. Cisplatin plus gemcitabine began on day 22 and was administered for four cycles, followed by radical cystectomy with lymph node dissection within 4 to 8 weeks after completion of chemotherapy. Adjuvant atezolizumab was planned only for patients with >ypT1 ypN0 after surgery.
The primary endpoint was centrally reviewed pathologic complete response, defined as ypT0 ypN0. Secondary endpoints included pathologic overall response, event-free survival, overall survival, recurrence-free survival after R0 resection, treatment feasibility, and safety.
Read more about Success Rate of Immunotherapy for Bladder Cancer on OncoDaily.
Patient Population and Treatment Exposure
Between April 2022 and April 2025, 47 patients were included in the full analysis set. The median follow-up for the primary analysis was 16.7 months, with a data cutoff of March 25, 2026.
Among the 47 patients, 40 underwent radical cystectomy with lymph node dissection. Seven patients did not undergo surgery: six declined and one was considered unfit for surgery.
Intravesical rBCG was administered to 95% of patients, and 37 patients received all three planned rBCG doses. Four doses of atezolizumab were completed by 43 patients. Four platinum administrations were also received by 43 patients, including patients who remained on cisplatin and those who switched to carboplatin.
Key Results
Centrally reviewed pathologic complete response was achieved in 27 of 40 resected patients, corresponding to 68%. The one-sided 95% confidence interval lower boundary was 53%, and the trial rejected the null hypothesis for the primary endpoint.
Pathologic overall response, defined as downstaging to ≤ypT1 ypN0, was reported in 33 of 40 resected patients, corresponding to 83%.
At the time of the analysis, 7 patients had experienced an event. The 12-month event-free survival rate was 90%, with a 95% confidence interval of 76% to 96%. The 12-month overall survival rate was 96%, with a 95% confidence interval of 84% to 99%.
R0 resection was achieved in 39 of 40 resected patients. Among patients who had R0 resection, 12-month recurrence-free survival was 92%, with a 95% confidence interval of 77% to 97%.
Complete clinical complete remission assessment was available for 25 of 40 resected patients. Among these patients, 12 achieved clinical complete remission after neoadjuvant treatment. Of those 12 patients, 11 had confirmed pathologic complete response. The positive predictive value of clinical complete remission was 92%, while the negative predictive value was 69%.
Safety
All patients experienced at least one treatment-related adverse event. No grade 5 adverse event was attributed to neoadjuvant therapy. One fatal treatment-related adverse event was reported in relation to surgery.
Treatment-related adverse events of any grade, grade 3, and grade 4 occurred in:
- rBCG: 42%, 9%, and 0%
- Atezolizumab: 55%, 15%, and 2%
- Chemotherapy: 96%, 38%, and 17%
For neoadjuvant therapy overall, grade 3 treatment-related adverse events occurred in 45% of patients and grade 4 events occurred in 17%. No grade 4 or grade 5 adverse events or unexpected toxicities were attributed to intravesical rBCG, and no systemic BCG dissemination was reported.
Comparison With SAKK 06/17
A preplanned propensity score–weighted comparison was performed with SAKK 06/17, a predecessor trial that used a similar chemoimmunotherapy backbone but did not include intravesical rBCG.
In the weighted comparison, pathologic complete response was 67% in SAKK 06/19 compared with 32% in SAKK 06/17. Pathologic overall response was 83% in SAKK 06/19 compared with 59% in SAKK 06/17. The comparison favored SAKK 06/19, but the trial design does not allow the benefit to be attributed definitively to rBCG alone.
Biomarker Findings
Exploratory cytokine and chemokine profiling showed early treatment-associated immune changes. After rBCG and atezolizumab, several immune mediators linked to lymphocyte recruitment and cytotoxic activity were preferentially induced, including CXCL9, CCL19, CXCL13, perforin, and granzyme B.
In comparison, samples from SAKK 06/17 showed a stronger pattern of chemokines associated with myeloid cell recruitment after chemotherapy. These biomarker findings support further investigation but remain exploratory because of the limited sample size and the possibility that changes could be related to atezolizumab rather than rBCG alone.
Study Limitations
The results should be interpreted carefully because SAKK 06/19 was a small, single-arm phase 2 trial with short follow-up. The analysis of pathologic response included only patients who underwent surgery, and 15% of patients did not proceed to surgery.
The study also lacked a randomized control arm, which limits conclusions about the specific contribution of rBCG. The comparison with SAKK 06/17 was preplanned and weighted, but it remained an intertrial comparison and included different immune checkpoint inhibitors. Longer follow-up is needed to determine whether the high pathologic complete response rate translates into durable disease control.
Takeaway
SAKK 06/19 suggests that adding intravesical recombinant BCG to perioperative chemoimmunotherapy may be a feasible strategy in operable muscle-invasive bladder cancer, with high pathologic complete and overall response rates and no systemic BCG dissemination reported. However, because this was a small, single-arm phase 2 trial with short follow-up, prospective randomized studies are needed to confirm the role of rBCG in this setting.
The full article is available in the Journal of Clinical Oncology.