Pembrolizumab + BCG

KEYNOTE-676 is Merck's bet that adding Keytruda to standard BCG therapy will improve outcomes in patients with high-risk non-muscle invasive bladder cancer (NMIBC), the earliest form of bladder cancer where the tumor sits in the bladder lining but has not invaded the muscle wall. Bacillus Calmette-Guérin (BCG), a live weakened tuberculosis bacterium instilled directly into the bladder, has been standard-of-care for 40 years. Roughly 30 to 50% of patients fail BCG within two years. The Phase 3 trial (NCT03711032) enrolls two cohorts: patients whose disease came back after BCG (persistent or recurrent) and patients who never had BCG (BCG-naive) [1]. The primary endpoint is complete response rate in patients with carcinoma in situ (CIS) at baseline, with event-free survival as a key secondary endpoint [1][8]. CIS is a flat, high-grade cancer pattern confined to the bladder lining surface and is the pattern that drives the most urgent treatment decisions in NMIBC because it is invisible to standard tumor resection and prone to progression. Pembrolizumab is already approved for a narrower slice of this population, BCG-unresponsive CIS patients who refuse or are ineligible for cystectomy, based on the single-arm KEYNOTE-057 Phase 2 [3]. KEYNOTE-676 is the lifecycle-management play to push Keytruda deeper into bladder cancer at the front line, where the patient pool is much larger. Merck reported roughly $29.5 billion in Keytruda revenue in 2024, and every earlier-line indication defended before US biologics loss of exclusivity in 2028 adds meaningful dollars [2].

Pembrolizumab is not a novel compound. Keytruda has been on the market since 2014 and carries approximately 40 FDA-approved indications as of mid-2026, including BCG-unresponsive high-risk NMIBC with CIS (accelerated approval in 2020 based on KEYNOTE-057) [3]. In September 2025 the FDA approved KEYTRUDA QLEX, a subcutaneous formulation co-formulated with berahyaluronidase alfa, across most existing solid-tumor indications for pembrolizumab, which shortens administration from a 30-minute infusion to a roughly one-minute injection [9]. KEYNOTE-676 is a Phase 3 combination and label-expansion trial pushing the drug into an earlier, much larger NMIBC population where BCG alone is standard [1][8]. No FDA breakthrough designation has been announced for the combination in this setting. Given pembrolizumab's existing NMIBC approval nearby, none is expected. The trial began dosing in late 2018 and remains active per ClinicalTrials.gov. Merck has not disclosed topline results as of July 2026 [2]. The BCG-persistent or recurrent cohort will likely read out before the BCG-naive cohort because event rates are higher and follow-up shorter. Regulatory strategy will hinge on whether Merck seeks a single combined label or separate approvals per cohort, which matters because the two populations have very different unmet need profiles and payer economics. The most immediate competitive comparator on timing is Pfizer's CREST trial of subcutaneous sasanlimab plus BCG in BCG-naive HR-NMIBC, which posted positive Phase 3 results in 2024 and is under regulatory review.

PD-1 is a brake pedal on T cells. When a T cell recognizes something suspicious like a tumor, the immune system also engages this brake to prevent the response from spiraling into autoimmunity. Cancer cells exploit that by displaying PD-L1, the molecule that presses the PD-1 brake, telling attacking T cells to stand down. Pembrolizumab is a monoclonal antibody that blocks PD-1 so the brake stays off and T cells keep killing tumor cells [4]. BCG works through an older, dirtier mechanism. Instilling live weakened tuberculosis bacteria into the bladder provokes a massive local inflammatory response that recruits innate and adaptive immune cells and, somehow, redirects them onto bladder cancer cells. Nobody fully understands why BCG works after 40 years of use, but it does, and it remains the only NMIBC therapy with an unambiguous long-term recurrence-free survival benefit [5]. The combination bet is straightforward. BCG floods the bladder with activated immune cells; pembrolizumab prevents those cells from being switched off by tumor-expressed PD-L1. Preclinical work supports this rationale, and PD-L1 upregulation is documented in BCG-exposed bladder tumors. Mechanism validation is strong for each drug individually. The combination is a reasonable extrapolation, not proven biology. The core question is whether checkpoint blockade adds enough incremental benefit on top of BCG to justify the cost, monitoring burden, and toxicity of a systemic antibody.

KEYNOTE-676 is a randomized Phase 3 trial (NCT03711032) with two cohorts running in parallel [1][8]. Cohort A enrolls patients with high-risk NMIBC that persisted or recurred after BCG induction and randomizes them to pembrolizumab plus BCG maintenance versus BCG maintenance alone. Cohort B enrolls BCG-naive high-risk NMIBC patients and randomizes them to pembrolizumab plus BCG versus BCG monotherapy. Total planned enrollment is roughly 975 patients across both cohorts. The primary endpoint is complete response (CR) rate in patients with CIS at baseline, matching the endpoint used for KEYNOTE-057's accelerated approval. Response is assessed by local cystoscopy and blinded independent central review of urine cytology and biopsy every 12 weeks for years one to two, then every 24 weeks. Key secondary endpoints include event-free survival (defined as high-grade recurrence, progression to muscle-invasive disease, cystectomy, or death from any cause), overall survival, safety, and quality of life. Using CR rate rather than EFS keeps the regulatory bar consistent with prior pembrolizumab NMIBC approvals but limits the primary readout to the CIS subgroup, which is a fraction of total enrollment. EFS in the full high-risk population is what will ultimately drive practice change and payer coverage, and that is a secondary and follow-on analysis. The persistent or recurrent cohort will drive the first regulatory conversation. A meaningful practical confounder is the recurring global BCG shortage that has forced dose-splitting and treatment substitutions in real-world NMIBC care since 2019, which can compress control-arm outcomes and distort observed relative benefit if trial supply is protected while real-world supply is not [8].

Our model estimates a 50% chance this drug is eventually approved. It starts from the historical base rate for Phase 3 drugs in this area (about 48%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by more secondary endpoints than usual, larger-than-typical enrollment for this phase, and its light or open-label blinding; it is held back by weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk is the biggest concern. Pembrolizumab worked in KEYNOTE-057 in a specific, more advanced group (BCG-unresponsive CIS refusing cystectomy) with a 41% CR rate at three months. Extrapolating benefit into earlier-stage disease that is already highly treatable with BCG monotherapy is not guaranteed. NMIBC event trajectories are slower than metastatic cancer, so checkpoint effect sizes may be diluted by long-tail recurrences. Safety risk exists but is manageable. Pembrolizumab has a decade-characterized immune-related adverse event profile including colitis, pneumonitis, and endocrinopathies. BCG toxicities including cystitis and BCG-osis (rare but serious systemic BCG infection requiring prolonged antimycobacterial antibiotic treatment) are also well described. Overlapping bladder inflammation will likely increase dosing interruptions and treatment discontinuations, which can bias survival curves if imbalanced between arms. Supply risk is real. Recurring global BCG shortages driven by manufacturing bottlenecks in the small number of BCG producers have disrupted NMIBC care for years and remain a plausible confounder for enrollment velocity and real-world control-arm performance. Execution risk is moderate. Two-cohort trials in the same operational shell can lose enrollment velocity when investigators triage patients. Commercial risk is the sleeper issue but softens with subcutaneous pembrolizumab. NMIBC is a urologist-driven disease, not a medical oncology disease. Historically, adding a $150,000+ per year systemic IV antibody to a locally delivered generic treatment has been a major workflow change for urology practices that mostly do not run infusion centers. KEYTRUDA QLEX subcutaneous dosing, given in about one minute, partially mitigates this by making in-office administration more plausible, though drug cost, monitoring for immune-related adverse events, and reimbursement pathways remain [6][9]. Payers will still demand meaningful benefit, not marginal separation. Competition is thickening: TAR-200 (Johnson & Johnson intravesical gemcitabine pretzel, filed for BCG-unresponsive HR-NMIBC), nadofaragene firadenovec (Ferring gene therapy, approved 2022), cretostimogene grenadenorepvec (CG Oncology BOND-003), and sasanlimab (Pfizer CREST) [7].

Worth watching, with focus on the BCG-persistent or recurrent cohort where readout comes first and event rates are highest. The primary signal to look for is a CR rate in the CIS subgroup clearly above the 41% KEYNOTE-057 bar, ideally with concordant early event-free survival separation, a supplemental BLA (a regulatory filing that adds a new indication to an already-approved drug) is the near-term commercial vehicle. Anything with an EFS hazard ratio above 0.85 is a likely failure to shift practice even if CR rate reads positive. Natural check-back points are ASCO GU January 2027 and ASCO May 2027, the venues Merck typically uses for major bladder cancer updates. This timing estimate assumes enrollment is complete or near-complete; Merck has not publicly disclosed enrollment status as of July 2026, and the estimated primary completion date on ClinicalTrials.gov has slipped historically, so 2027-01 should be treated as a leading estimate rather than a firm date [1]. Strategic motivation for Merck is strong: Keytruda faces US biologics loss of exclusivity in 2028, and every defended or expanded indication chips away at biosimilar erosion of a $29.5 billion product [2]. Bladder cancer front line is a several hundred million dollar opportunity if the label lands and urologists adopt, and subcutaneous Keytruda QLEX materially improves the odds of that workflow fit [9]. If KEYNOTE-676 fails, expect Merck to lean harder on Keytruda's muscle-invasive bladder story (KEYNOTE-B15, KEYNOTE-905, KEYNOTE-992 perioperative and neoadjuvant trials) rather than retreat. Companion assets to track are Pfizer sasanlimab (already positive in CREST), Johnson & Johnson TAR-200 (regulatory decision pending), CG Oncology cretostimogene grenadenorepvec, and Ferring nadofaragene firadenovec follow-ons [7]. The broader question, whether systemic checkpoint blockade adds enough to intravesical therapy to justify its cost and toxicity in early bladder cancer, will be answered across this cluster of trials over the next 18 months.