SAKK 06/19: rBCG added to chemo-IO achieves high pCR in MIBC

Adding intravesical recombinant BCG (rBCG; VPM1002BC) to perioperative chemoimmunotherapy yielded a pathological complete response (pCR) rate of 68% in patients with muscle-invasive bladder cancer (MIBC), meeting the primary end point of the phase 2 SAKK 06/19 trial (NCT04630730), according to results presented at the 2026 American Society of Clinical Oncology Annual Meeting.¹

Richard Cathomas, MD, head of the division of oncology/haematology at Kantonsspital Graubünden in Chur, Switzerland, presented the findings on behalf of the Swiss Cancer Institute Network. Cathomas characterized the results as evidence that the combination is feasible, tolerable, and carries promising efficacy, while calling for further evaluation in randomized trials.

Background and rationale of SAKK 06/19

Perioperative immunotherapy has been established as feasible in MIBC, and the phase 3 NIAGARA trial (NCT03732677) demonstrated improved event-free survival (EFS) and overall survival (OS) with a similar chemoimmunotherapy regimen compared with chemotherapy alone.^1,2 However, pCR rates with current regimens remain below 40%, representing an unmet clinical need.

Intravesical BCG has been used for decades in non–muscle invasive bladder cancer (NMIBC), where it induces local inflammation and activates both innate and adaptive immune responses. The novel rBCG is genetically modified through the exchange of the urease C gene with the hemolysin listeriolysin gene from Listeria monocytogenes, enhancing immunogenicity via antigen egress to promote adaptive immune cross-presentation and innate immune activation via double-stranded DNA release. rBCG is also engineered for improved safety through more rapid elimination by autophagy, reducing organism persistence.^1,3

A prior SAKK study (SAKK 06/14) of rBCG in BCG-exposed NMIBC demonstrated safety and promising efficacy, supporting the hypothesis that induction therapy with intravesical rBCG would improve the efficacy of systemic perioperative chem-immunotherapy in operable MIBC.^1,4

Study design

SAKK 06/19 was a prospective, single-arm, open-label, multicenter phase 2 trial enrolling patients with urothelial carcinoma (UC) of the bladder (clinical T2–T4a, cN0/N1), World Health Organization (WHO) performance status 0 to 1, cisplatin eligibility (glomerular filtration rate >50 mL/min), and no prior intravesical BCG. UC subtypes were allowed (excluding pure squamous or glandular histology); patients with a small cell component were excluded.

Treatment consisted of 3 phases. In the neoadjuvant phase, patients received 3 weekly intravesical rBCG instillations (days 1, 8, and 15), atezolizumab (Tecentriq) every 3 weeks for 4 cycles, beginning on day 1, and cisplatin/gemcitabine every 3 weeks for 4 cycles, beginning on day 22. Radical cystectomy with extended pelvic lymph node dissection (RC-LND) was performed 4 to 8 weeks after the end of chemotherapy. Adjuvant atezolizumab (13 cycles every 3 weeks) was administered only to patients with residual muscle-invasive disease (≥ypT2) or lymph node involvement (ypN+). Patients achieving ≤ypT1 ypN0 were transferred directly to follow-up.

The primary end point was pCR (ypT0 ypN0) at cystectomy per central review. Secondary end points included pathological response (PaR, ≤ypT1 ypN0), EFS (defined as time from treatment start to progression leading to inoperability, local recurrence, distant metastasis, or death), OS, feasibility, and safety. The trial used a Simon's minimax 2-stage design (H0: pCR ≤35%; H1: pCR ≥55%; type I error 5%, power 80%), requiring 39 resected patients; 46 total patients were planned to account for a 15% dropout rate.

Patient population

A total of 50 patients were screened; 47 were enrolled between April 2022 and April 2025. Accrual was suspended for approximately 10 months (October 2022 to August 2023) due to rBCG being found out of specification, necessitating remanufacturing. Median age was 67 years (range 24–78); 81% of patients were male. WHO performance status was 0 in 79% of patients. Clinical T stage was cT2 in 60%, cT3 in 32%, and cT4a in 8%; 11% had cN1 disease. Macroscopically incomplete transurethral resection of bladder tumor was reported in 64% of patients. Pure UC histology was present in 77%; 23% had UC with a subtype.

All 47 patients completed neoadjuvant treatment. Adherence was high: 79% received all 3 planned rBCG doses, 92% received all 4 atezolizumab doses, 92% received all 4 platinum doses (81% received full cisplatin; 19% were switched to carboplatin), and 87% received at least 7 of 8 planned gemcitabine doses. Seven patients did not undergo cystectomy: 6 refused (including 1 withdrawal), and 1 was deemed unfit for surgery. Of the 40 patients who underwent RC-LND, 31 did not receive adjuvant atezolizumab—30 per protocol (≤ypT1 ypN0) and 1 per physician decision.

Efficacy outcomes

Among the 40 patients in the cystectomy set, 27 achieved pCR (ypT0 ypN0), for a pCR rate of 68% (one-sided 95% CI lower bound 53%; P < .0001), rejecting the null hypothesis. PaR (≤ypT1 ypN0) was achieved by 33 patients (83%). Among non-pCR patients, residual disease included ypTa in 1 patient (3%), ypTis in 3 (8%), and ypT1 in 2 (5%). Non-PaR disease (≥ypT2 and/or ypN+) was present in 7 patients (18%), comprising ≥ypT2 ypN0 in 2 (5%), ≥ypT2 ypN+ in 4 (10%), and ypT0 ypN+ in 1 (3%).

"Twenty-seven of 40 patients achieved a pCR amounting to a rate of 68%. Therefore, the null hypothesis could be rejected," Cathomas said.

Among 25 patients with evaluable presurgery clinical complete response (cCR) data, cCR—defined by CT thorax/abdomen/pelvis, MRI bladder, cystoscopy, and urine cytology—carried a positive predictive value for pCR of 92% and a negative predictive value of 69%.

With a median follow-up of 16.7 months in all 47 enrolled patients, the 12-month EFS rate was 90% (95% CI, 76%–96%), and the 12-month OS rate was 96% (95% CI, 84%–99%).

Comparison with SAKK 06/17

A preplanned, propensity score-weighted (PSW) comparison was conducted between SAKK 06/19 and the predecessor SAKK 06/17 trial, which evaluated cisplatin/gemcitabine plus durvalumab (without intravesical rBCG) in the same centers under very similar eligibility criteria. In the PSW multivariate analysis, the odds ratio for pCR in SAKK 06/19 vs SAKK 06/17 was 5.13 (95% CI, 2.63–10.02; P < .001). The pCR rate was 67% in SAKK 06/19 vs 32% in SAKK 06/17. Cathomas noted that the increase in pCR was driven predominantly by a reduction in residual muscle-invasive disease rather than non–muscle invasive disease.

"When comparing the 2 studies, it is clear that the pCR rate is much higher in the current SAKK 06/19 study," Cathomas said.

Safety

Treatment-related adverse events (TRAEs) were assessed using NCI CTCAE version 5. rBCG-related adverse events were generally mild: grade 1 in 18%, grade 2 in 16%, and grade 3 in 9% of patients; no grade 4 rBCG-related events occurred, and no systemic BCG infection was reported. Atezolizumab-related grade 3 and 4 events occurred in 15% and 2% of patients, respectively. Chemotherapy was the primary driver of grade 3/4 toxicity, with grade 3 events in 38% and grade 4 events in 17% of patients.

The most common TRAEs of any grade included neutrophil count decrease (55%), nausea (47%), fatigue (45%), renal impairment (44%), platelet count decrease (33%), and urinary tract infection (32%). Grade 3/4 TRAEs were predominantly chemotherapy-related, including neutropenia, thrombocytopenia, urinary tract infections, anemia, thromboembolic events, and renal impairment.

Limitations and conclusions

Cathomas identified several limitations of the trial, including its nonrandomized, single-arm design, small sample size, and short follow-up at the time of primary analysis. Potential selection bias was acknowledged in the context of a complex interdisciplinary regimen, and the 15% nonsurgery rate introduces additional bias risk in interpreting response data. With respect to the comparison with SAKK 06/17, Cathomas noted that different checkpoint inhibitors were used (atezolizumab vs durvalumab [Imfinzi]) and that potential unmeasured confounders remain despite propensity score weighting.

In his concluding remarks, Cathomas stated that the combination is feasible and tolerable, that the preplanned comparison to SAKK 06/17 supports a potential impact for intravesical rBCG, and that further investigation of intravesical rBCG in combination with next-generation systemic regimens in randomized trials is warranted.

DISCLOSURES: Cathomas reported institutional honoraria from Astellas Pharma, AstraZeneca, Johnson & Johnson/Janssen, and Merck KGaA; a consulting or advisory role with Astellas Pharma (institutional), AstraZeneca (institutional), Bayer (institutional), Bristol Myers Squibb (institutional), Ipsen (institutional), Johnson & Johnson/Janssen, MSD Oncology (institutional), Pfizer (institutional), and Roche (institutional); and travel expenses from AstraZeneca (institutional).

REFERENCES

1. Cathomas R, Petrausch U, Hayoz S, et al. Intravesical recombinant BCG combined with chemo-immunotherapy (chemo-IO) as perioperative therapy for patients with muscle-invasive bladder cancer (MIBC): primary analysis of SAKK 06/19. J Clin Oncol. 2026;44(suppl 16):4503. doi:10.1200/JCO.2026.44.16_suppl.4503

2. Powles T, Catto JWF, Galsky MD, et al; NIAGARA Investigators. Perioperative durvalumab with neoadjuvant chemotherapy in operable bladder cancer. N Engl J Med. 2024;391(19):1773-1786. doi:10.1056/NEJMoa2404877

3. Grode L, Seiler P, Baumann S, et al. Increased vaccine efficacy against tuberculosis of recombinant Mycobacterium bovis bacille Calmette-Guérin mutants that secrete listeriolysin. J Clin Invest. 2005;115(9):2472-2479. doi:10.1172/JCI23142

4. Rentsch CA, Birkhauser FD, Biot C, et al. Intravesical recombinant BCG (VPM1002BC) in BCG-exposed patients with non-muscle invasive bladder cancer: a phase 2 study. Eur Urol Oncol. 2022;5:195-202. doi:10.1016/j.euo.2021.08.009