Intermediate and high-risk non-muscle-invasive bladder cancer: an overview of epidemiology, burden, and unmet needs

Bladder cancer ranks among the most common cancers globally. At diagnosis, 75% of patients have non-muscle-invasive bladder cancer (NMIBC). Patients with low-risk NMIBC have a good prognosis, but recurrence and progression rates remain high in intermediate- and high-risk NMIBC, despite the decades-long availability of effective treatments for NMIBC such as intravesical Bacillus Calmette-Guérin (BCG). The present review provides an overview of NMIBC, including its burden and treatment options, and then reviews aspects that counteract the successful treatment of NMIBC, referred to as unmet treatment needs. The scale and reasons for each unmet need are described based on a comprehensive review of the literature, including insufficient adherence to treatment guidelines by physicians because of insufficient knowledge, training, or access to certain therapy options. Low rates of lifestyle changes and treatment completion by patients, due to BCG shortages or toxicities and adverse events as well as their impact on social activities, represent additional areas of potential improvement. Highly heterogeneous evidence for the effectiveness and safety of some treatments limits the comparability of results across studies. As a result, efforts are underway to standardize treatment schedules for BCG, but intravesical chemotherapy schedules remain unstandardized. In addition, risk-scoring models often perform unsatisfactorily due to significant differences between derivation and real-world cohorts. Reporting in clinical trials suffers from a lack of consistent outcomes reporting in bladder cancer clinical trials, paired with an under-representation of racial and ethnic minorities in many trials.


High-Quality Treatment Guidelines
Treatment guidelines considered to be "strongly recommended" by Maisch et al. (1) are summarised in Table S1 or, in the case of the National Institute for Health and Care Excellence, in Table S2. Additional guidelines are summarised in Table S2.  Treatment in a community cancer center is associated with a higher risk of non-guidelinecompliant therapy than in an academic center.

Reasons for/factors influencing non-compliance
Choo et al. (11) • Second TURBT if no detrusor muscle in the first TURBT specimen is more likely in academic teaching than in private or non-teaching hospitals and more likely for surgeons with shorter than with longer practice (no statistically significant difference for either group regarding the second TURBT for T1 grade or high-grade disease) • Single immediate chemotherapy instillation more likely in academic teaching than in private or non-teaching hospitals for T1 and high-grade disease (no statistically significant difference for other indications) • Reported country-specific differences (between Japan, Korea, and Taiwan) that could not readily be explained Jeglinschi et al. (13) Guideline deviation based on a decision by treating urologist without reasons specified (of all guideline deviations in the risk group): • Intermediate-risk patients not receiving second TURBT: 100% (one patient) • High-risk patients not receiving second TURBT: 65.0% • Intermediate-risk patients not starting intravesical therapy: 79.0% • High-risk patients not starting intravesical therapy: 53.5% • Very high-risk patients not receiving radical cystectomy: 18.2% Matulay et al. (21), online survey of 121 urologists in the US • Higher rates of guideline compliance in urologic oncologists versus non-urologic oncologists and in fellowship-trained versus non-fellowship-trained urologists • Higher rates of guideline compliance among physicians based in academic relative to hospital employment and among physicians with shorter relative to longer experience

Study, study design Reasons for/factors influencing non-compliance
Tobert et al. (18) Guideline compliance is more likely if treated in an academic cancer center relative to nonacademic, non-cancer centers

Lack of access and economic barriers
Balakrishnan et al. (20) Lack of insurance is associated with a higher risk of non-guideline-compliant therapy than being privately or Medicare-insured Living in areas with lower adult educational attainment is associated with a higher risk of nonguidelines-compliant therapy Choo et al. (11) BCG shortage is the main reason for not giving maintenance BCG, particularly in Taiwan (less so in Japan, which has a more stable supply) Jeglinschi et al. (13) Guideline deviation due to lack of resources (of all guideline deviations): • High-risk patients not receiving second TURBT: 2.3% Wang et al. (19) • 75% of guideline deviations for BCG are due to inaccessible BCG • 50.4% of indicated second TURBTs, 40.7% of indicated BCG therapies, and 29.2% of indicated radical cystectomies were rejected by patients for economic reasons

Fear of complications or patient comorbidities
Jeglinschi et al. (13) Guideline deviation due to poor patient health status (of all guideline deviations in risk group): • High-risk patients not receiving second TURBT: 20.0% • High-risk patients not starting intravesical therapy: 23.3% • Very high-risk patients not receiving second TURBT: 60.0% • Very high-risk patients not receiving radical cystectomy: 63.6% Wang et al. (19) • 37.6% of indicated second TURBTs were not performed because the urologist was concerned about the risk of side effects; in 70.6%, the patient expressed such concerns • 26.0% of indicated BCG therapies were not performed because the urologist was concerned about the risk of side effects; in 62.2%, patients expressed such concerns Reasons for/factors influencing non-compliance • 38.1% of indicated radical cystectomies were rejected by patients concerned about side effects, while 57.7% were rejected by patients out of concerns for decreases in quality of life (47.4% rejected due to "personal reasons") Jeglinschi et al. (13) Poor patient compliance is the reason for:

Treatment Discontinuation in Patients with NMIBC
• High-risk patients not starting BCG: 21% • High-risk patients not finishing BCG maintenance: 35% • High-risk patients not undergoing second TURBT: 15% • Very high-risk patients not undergoing radical cystectomy: 18%

6
Validation Results for Risk Scoring Models • CUETO model underestimates 1-year recurrence and progression risks and overestimates 5-year progression risks (validated in two cohorts from Spanish and Polish institutions) (36) • CUETO model has poor discrimination for recurrence and progression and overestimates risks for either outcome in high-risk patients, including in BCG-treated patients (validated in an international multi-center study) (37) • CUETO model can stratify risks but has poor discrimination for predicting clinical events (validated in a multi-institutional cohort) (38) EORTC (39) • EORTC model underestimates 1-and 5-year recurrence and progression risks (validated in Swedish cohort) (35) • EORTC model has poor discrimination for recurrence and progression and overestimates risks for either outcome in high-risk patients, including in BCG-treated patients (validated in an international multi-center study) (37) • EORTC model stratifies recurrence and progression risk but has reduced discriminative ability for progression in patients treated with BCG (validated in CUETO data) (40) • EORTC model performs better than CUETO and EAU pre-2021 models for predicting recurrence, progression, and mortality, but overall performance is modest (validated in a single-institution cohort from Poland) (41)

Model/risk table Validation findings
EAU 2021 model (42) • EAU 2021 model underestimates 1-and 5-year recurrence and progression risks (validated in Swedish cohort) (35) • EUA 2021 model can stratify risks but has poor discrimination for predicting clinical events (validated in a multi-institutional cohort) (38) • EAU 2021 prognostic risk factor groups stratify progression risk appropriately but overestimate progression risk in patients receiving BCG (validated single-institution cohort from the US) (43) • EAU 2021 model introduces a very high-risk group that allows identifying patients more likely to progress but overall model accuracy is limited in patients with repeat transurethral resection of the bladder and BCG (assessed in a mult-institutional cohort from Italy) (44)