Key points

Introduction

Non–muscle invasive bladder cancer (NMIBC) remains a challenging disease to treat, with Ta, T1, and carcinoma in situ (CIS) stages comprising nearly 70% of all new diagnoses of bladder cancer. ^, High-risk NMIBC (defined by the American Urological Association/Society for Urologic Oncology [SUO] joint guidelines as T1 high-grade disease, any recurrent Ta high grade(HG) disease, Ta HG disease >3 cm or disease that is multifocal, presence of CIS, variant histology, lymphovascular invasion, high-grade prostatic urethral involvement, or prior bacillus Calmette-Guérin [BCG] failure) is aggressive and marked by frequent disease recurrences and risk of stage progression, contributing to high surveillance, treatment-related costs, and patient anxiety. Despite this, the mainstay of high-risk NMIBC clinical management remains transurethral resection of bladder tumor followed by intravesical instillations of BCG.

Estimates for the rates of disease recurrence following BCG therapy have ranged from 35% to 40%, but true recurrence patterns are difficult to ascertain because of the different definitions of BCG failure and confounding conferred by patients receiving variable total doses of BCG. Nevertheless, it is reasonable to assume that most patients with NMIBC will eventually develop recurrent disease, and up to 35% of these patients will die of their bladder cancer. The gold standard for BCG salvage therapy in high-risk NMIBC is to proceed with a radical cystectomy with bilateral pelvic lymphadenectomy, but this procedure is associated with significant morbidity and not all patients are appropriate surgical candidates. To date, the only alternative Food and Drug Administration (FDA)-approved therapy for BCG-refractory CIS is valrubicin, but this option is associated with a complete response (CR) rate of only 10% at 12 months. There remains a significant unmet need for alternative effective salvage treatment options as second-line therapy for patients with NMIBC facing cystectomy.

Defining bacillus Calmette-Guérin salvage therapy

In 2012, the SUO along with representatives from the FDA met to discuss a registration strategy for trials being considered for the salvage treatment of patients with NMIBC. It is worth noting that at this point, patients who had “failed” BCG were referred to by terms that had vague or imprecise definitions, such as “BCG-refractory” or “BCG-resistant.” The definition of these terms was based primarily on the treating physicians’ interpretation of their patients’ clinical status. At the Genitourinary Cancers Symposium in 2015, a task force helped clarify the disease state by formally establishing the term “BCG-unresponsive.” Patients who were classified as BCG-unresponsive were defined as those who had been treated with adequate BCG but had recurrent disease. Importantly, “adequate BCG” was defined as having received at least five of six induction instillations of BCG and at least two of three instillations of maintenance BCG. The 2018 FDA statement on BCG-unresponsive NMIBC specifically defines BCG-unresponsive as patients who have:

• •
  

  Persistent or recurrent CIS alone or with Ta/T1 disease within 12 months of adequate BCG therapy

  
  
• •
  

  Recurrent high-grade Ta/T1 disease within 6 months of adequate BCG therapy

  
  
• •
  

  T1 high-grade disease at the first evaluation following an induction course of BCG ^,

Addressing the problem: establishing a trial registration strategy

Of crucial concern was the lack of treatment options available for patients with BCG-unresponsive disease, and the dearth of clinical trials to address the unmet need in this space. Since 1959, only four agents had been approved for use in NMIBC. These included thiotepa for superficial papillary carcinoma of the urinary bladder (1959), doxorubicin (1974), BCG (1989) for CIS and prophylaxis of recurrent papillary tumors, and valrubicin (1998) for BCG-refractory CIS in patients for whom immediate cystectomy would be unacceptable. ^,

In 2012, under the aegis of the SUO Winter meeting in Bethesda, Maryland, a discussion took place among clinicians, industry partners, and the FDA to negotiate the terms for a registration strategy for BCG-unresponsive NMIBC. Among the issues raised included determining the most appropriate trial design type (single- vs double-arm design), clinically meaningful end points (recurrence-free survival or CR), and defining what constituted an appropriate patient mix to study (ie, a mixed population of Ta HG, T1, and CIS, vs cohorts of patients who only had CIS). The latter distinction was particularly important given that treatment of papillary bladder tumors is in essence adjuvant therapy following transurethral resection of bladder tumor, whereas treatment of CIS is classified as primary therapy. The SUO Clinical Trials Consortium/FDA meeting in November 2012 led to a consensus that ideal trials in this space should be comprised of a mixed population of HG tumors (CIS, Ta, and T1 tumors) with a minimum of 50% having some element of CIS. Single-arm trials were favored over two-arm trials, given that there currently exists no acceptable comparator treatment of patients who are BCG-unresponsive (aside from radical cystectomy, to which randomization would be challenging and not equivalent to medical intravesical therapy). Most on the panel considered 12 months to be an acceptable minimum duration for a clinically meaningful response. The SUO Clinical Trials Consortium bladder committee agreed on a 25% freedom from HG recurrence or CR as a meaningful clinical end point for success. A subsequent meeting between bladder cancer experts at the American Urological Association with the FDA in May 2013 provided further consensus in support of single-arm trial designs with a mixed population, acknowledging that the treatment intent for papillary tumors (adjuvant) and CIS (treatment) are distinct. This panel initially recommended a goal CR of 40% to 50% at 6 months with a durable response rate of 30% at 18 to 24 months for CIS. Later these metrics were recognized as being overly ambitious, and the bar was lowered slightly to provide for an incremental but meaningful improvement over valrubicin.

Validating the bacillus Calmette-Guérin–unresponsive definition and the hope for future therapeutic options

The BCG-unresponsive definition was evaluated at the MD Anderson Cancer Center in a retrospective cohort study of 83 patients with NMIBC, 55 of whom were found to have satisfied the formal FDA definition of BCG-unresponsive disease, with the remaining 28 patients developing recurrence after induction therapy alone. Although there were no differences between primary and recurrent tumor pathology of the two groups before or after receiving BCG, meeting the definition of BCG-unresponsive was independently predictive of worse cystectomy-free survival (hazard ratio, 3.85; 95% confidence interval, 1.49–10.0; P = .006). Additionally, BCG-unresponsive patients had significantly worse high-grade recurrence free survival (hazard ratio, 6.25; 95% confidence interval, 2.27–16.67; P <.001). This study in part validated the futility of using currently available intravesical agents for patients who reached the BCG-unresponsive definition, and it further underscored the urgent need for new therapies in this area. To this end, it is promising that there are at least eight ongoing clinical trials in the BCG-unresponsive space that are actively recruiting patients, including those considering systemic and intravesical treatment options. Additional trials are being planned at the time of this publication.

In summary, BCG-unresponsive disease remains a challenging disease to treat with limited therapeutic options aside from the gold-standard of radical cystectomy with pelvic lymphadenectomy. Participation in clinical trials in situations where it is safe and appropriate remain critical, particularly for patients who are not suitable candidates for surgery. Additionally, the dissemination, understanding, and correct use of BCG-unresponsive terminology is important, particularly when deciding on which patients may be best suited for clinical trial enrollment and in identifying patients for whom additional BCG therapy would be futile. This issue of Urologic Clinics seeks to highlight previous and ongoing efforts at developing alternative approaches to the management of intravesical treatment of high-risk BCG-unresponsive NMIBC, within the context of salvage therapy. The intermittent shortages of BCG that have affected our ability to treat this patient population makes the need for identification of alternatives even more acute. Current efforts suggest that several validated options could emerge for this challenging patient population in the near future.