Why has the prognosis for muscle-invasive bladder cancer not significantly improved after decades of therapeutic advancements?

KEYWORDS:

• Bladder cancer
• chemotherapy
• cystectomy
• immunotherapy
• prognosis
• targeted therapy

1. Introduction

Bladder cancer (BCa), mostly urothelial carcinoma (UCa), is often a highly aggressive disease. According to the Surveillance, Epidemiology, and End Results (SEER) database, the 5-year relative survival rate for those diagnosed with BCa at any stage in 1975–1977 was 72.3%, while that in more recent cases (2009–2015) was 78.3% [1]. Thus, the mortality rate appears to have improved non-significantly in the last several decades or has been even lowered in some countries or regions. The improvement rate is indeed much lower, compared with those in other urological malignancies such as prostate (i.e. 30% improvement) and renal (i.e. 25% improvement) cancers. Meanwhile, recent advances in BCa diagnosis include, but are not limited to, the availability of various urine biomarker assays [2] and new technologies [3], such as blue-light cystoscopy, narrow-band imaging, and positron emission tomography scan, although no screening test has been recommended for the early detection of BCa. This article summarizes and discusses available data indicating why BCa, especially muscle-invasive tumor, remains a lethal disease despite considerable therapeutic advancements.

2. Current status and recent advance in treatment

2.1. Cystectomy

Radical cystectomy is the current standard of treatment for muscle-invasive BCa with no distant metastasis. Importantly, no significant recent changes in oncologic outcomes in patients undergoing radical cystectomy have been reported in relatively large clinical trials. For instance, in a study involving 1488 cases undergoing radical cystectomy between 1980 and 2005, no survival improvement was noted, while confounding factors, such as lymph node involvement, were shown to have a clear negative impact [4].

Since its initial application in 2003, robot-assisted laparoscopic radical cystectomy has been adopted globally. However, most studies have shown no significant difference in the rate of positive surgical margin or subsequent recurrence/progression, as well as that of major complications or quality of life, between robot-assisted laparoscopic versus open cystectomies. However, the robotic approach appeared to reduce the risk of blood transfusions substantially and length of hospital stay slightly [5]. It is worthy to mention that pneumoperitoneum initiated during laparoscopic surgery may correlate with tumor seeding through various mechanisms, including impairing the immune response, assisting the migration of tumor emboli, and decreasing pH, all of which result in the promotion of cell implantation.

2.2. Trimodal therapy

Organ-sparing procedures have been established as alternatives to radical surgery. Trimodal therapy consisting of either transurethral resection of the bladder or partial cystectomy followed by radiotherapy with concurrent chemotherapy is currently considered to yield the best oncologic outcomes in select patients with muscle-invasive BCa. A recent meta-analysis demonstrated the findings in favor of radical cystectomy, with hazard ratios (95% confidence interval [CI]) of 1.39 (1.03–1.88) and 1.39 (1.20–1.59) for disease-specific survival (DSS) and overall survival (OS), respectively [6]. However, trimodal therapy in appropriately selected patients has been suggested to increase quality-adjusted life years, compared with radical cystectomy. Current treatment guidelines indicate that optimal patients for bladder preservation with trimodal therapy include those with T2NXM0 or T2N0M0 disease without hydronephrosis or extensive UCa in situ [7].

2.3. Systemic therapy

The efficacy of a combination treatment, MVAC (methotrexate/vinblastine/doxorubicin/cisplatin), in patients with BCa was first reported in the late 1980s. MVAC has thereafter been shown to provide a survival advantage over cisplatin alone (e.g. median survival in patients with metastatic BCa: 12.5 vs. 8.2 months [8]). In 2000, GC (gemcitabine/cisplatin), showing comparable anti-tumor activity and similar overall response to MVAC [9], was applied to the treatment of BCa. Both regimens have been widely used in neoadjuvant and adjuvant settings. Indeed, combined neoadjuvant chemotherapy and radical cystectomy have been shown to considerably improve the status of residual disease at surgery and long-term prognosis without increases in treatment-related morbidity and mortality. Carboplatin-based chemotherapy achieving inferior outcomes is also undergone, especially in those who are ineligible to cisplatin due to renal dysfunction and/or severe comorbidities. However, other chemotherapy regimens (that are more effective and/or less toxic) are not routinely used for BCa treatment.

Identification of molecules that play a key role in UCa progression has led to the development of targeted therapy. Several agents targeting specific pathways have indeed been approved for the clinical use by the Food and Drug Administration (FDA) in the US in 2016 or after, while others are still under clinical investigation (Table 1).

Table 1. Targeted therapy drugs approved by the FDA or currently in clinical trials for UCa.

Target / Agent	FDA approval
PD-1/PD-L1   Atezolizumab   Avelumab   Durvalumab   Nivolumab   Pembrolizumab   Toripalimab	May 2016 May 2017 May 2017 February 2017 May 2017 No
FGFR   Bemarituzumab   Erdafitinib   Infigratinib   Rogaratinib   Pemigatinib   Vofatamab	No April 2019 No No No No
VEGF   Axitinib   Cabozantinib   Ramucirumab	No No No
ErbB   Afatinib   Erlotinib   Trastuzumab	No No No
Multiple RTKs   Sitravatinib	No
Nectin-4   Enfortumab vedotin	No
mTOR   Everolimus   Nab-sirolimus	No No
CTLA4   Ipilimumab   Tremelimumab	No No
Proteasome   Ixazomib	No
IDO1   KHK2455	No
CDK4/6   Abemaciclib	No
LAG-3   TSR-033	No

RTK: receptor tyrosine kinase; mTOR: mammalian target of rapamycin; CTLA4: cytotoxic T-lymphocyte-associated protein 4; IDO1: indoleamine 2,3-dioxygenase 1; CDK: cyclin-dependent kinase; LAG-3: lymphocyte activation gene-3

Immune checkpoints serve to prevent autoimmunity, and their selective inhibitors have been shown to attack tumor cells via enhancing the immune response rather than direct cytotoxicity. Two FDA-approved antibodies against programmed cell death-1 (PD-1) and its ligand (PD-L1), atezolizumab and pembrolizumab, can be used in patients with advanced or metastatic BCa who are ineligible for cisplatin-based therapy as first-line treatment. Additional three PD-1/PD-L1 antibodies, avelumab, durvalumab, and nivolumab, were also approved for second-line treatment. In some studies, the superior survival benefit of immune checkpoint inhibitors over standard chemotherapy was observed [10].

Mutations, translocations, or dysregulated expression in the fibroblast growth factor receptor (FGFR) genes are known to involve BCa development and progression. Erdafitinib, a selective pan-FGFR kinase inhibitor, was also approved by the FDA in April 2019 as a second-line treatment for BCa. An open-label phase II study demonstrated a promising response rate of 40%, with 5.5 months of progression-free survival (PFS) and 13.8 months of OS (plus 30% of the responses maintained for more than 12 months) in previously treated patients with locally advanced or unresectable/metastatic BCa exhibiting FGFR alterations [11]. The efficacy of other FGFR inhibitors, such as infigratinib, rogaratinib, pemigatinib, and vofatamab, in advanced BCa has also been assessed in phase I–III trials.

Vascular endothelial growth factor (VEGF) is a key mediator of angiogenesis which is a promising target for anti-tumor therapy. A phase II trial investigating the efficacy of bevacizumab, a monoclonal antibody directed against circulating VEGF-A, in combination with GC as first-line therapy for metastatic UCa showed an overall response rate of 72% with 8.2 months of PFS and 19.1 months of OS [12]. However, a subsequent phase III trial (NCT00942331) failed to confirm the benefit of bevacizumab over GC alone for metastatic UCa (OS: GC+bevacizumab – 14.5 months vs. GC+placebo – 14.3 months). A phase III study of ramucirumab, a monoclonal antibody targeting VEGF receptor, plus docetaxel in UCa is ongoing (NCT02426125), while disappointing results have been obtained in clinical trials for UCa with other VEGF inhibitors, including sunitinib and pazopanib.

The ErbB family, consisting of ERBB1/EGFR, ERBB2/HER2, ERBB3, and ERBB4, is a class of receptor tyrosine kinases that are well known to regulate survival signaling in BCa cells. In a phase II study enrolling 23 patients with metastatic and cisplatin-refractory UCa, afatinib, an irreversible inhibitor of the ErbB family, significantly prolonged the median time to progression/discontinuation to 6.6 months in those with ERBB2/ERBB3 alterations (vs. 1.4 months without alterations) [13]. There is an ongoing phase II trial of afatinib monotherapy in patients with UCa carrying ERBB2/ERBB3 mutations or EGFR amplifications (NCT02780687). However, no pronounced benefit of HER2 inhibitors, such as trastuzumab and lapatinib, with or without combination with GC, has been observed in clinical trials involving UCa patients.

Enfortumab vedotin is an antibody-drug conjugate targeting Nectin-4, a member of the Nectin transmembrane adhesion protein family. In an ongoing phase II trial of enfortumab vedotin in patients with locally advanced or metastatic UCa previously treated with immune checkpoint inhibitors, the overall response rate was reported to be 44% (95% CI: 35.1–53.2%), including 12% of complete response (NCT03219333).

3. Conclusion and expert opinion

As discussed, despite some progress in managing muscle-invasive and/or metastatic BCa, its prognosis has not significantly improved in the last several decades. However, new drugs for targeted therapy, including those recently approved by the FDA, have shown to prolong DSS/OS by at least several months. It has also been documented that a considerable portion of cases achieves complete response by some of these drugs. Moreover, the results from ongoing trials are expected to provide further improvement in patient outcomes.

Why have these apparent advances not so far significantly changed patient outcomes? There are several issues that need to be addressed if 1) recent advancement has not been adopted or has been inappropriately adopted; 2) the populations in clinical trials reflect real-world patients; 3) there are reasons why some agents are active in phase II trials, but not in phase III; and 4) BCa has changed in its nature. Of note, only a subset of real-world patients has been reported to indeed receive systemic therapy, often in the last few months of their life [14]. Otherwise, patients eligible in clinical studies (particularly phase II trials in a single institution) likely had better performance status than real-world populations. Moreover, the qualities, such as data collection and protocol compliance, in phase III trials are not generally as good as those in phase II studies that are often conducted only in select institution(s). Meanwhile, in previous clinical trials, histological variants of urothelial carcinoma, whose behavior is generally more aggressive than conventional cancer, may have been excluded.

It is also of importance to develop strategies for overcoming resistance to currently available therapies. As one example, we have demonstrated preclinical evidence indicating that activation of androgen-mediated androgen receptor (AR) signaling is associated with the induction of not only urothelial tumorigenesis and tumor progression but also resistance to conventional therapies for BCa, including cisplatin treatment and irradiation [15]. We, therefore, anticipate that anti-androgenic agents function as sensitizers that enhance the effects of cisplatin or irradiation in BCa patients. A phase I/Ib trial of an AR antagonist, enzalutamide, combined with GC in patients with AR-positive metastatic BCa is conducted, preliminarily demonstrating the median PFS/OS of 7.68/10.59 months, respectively (NCT02300610).

Further optimization of treatment options, presumably based on molecular diagnostics, is required. In particular, it is crucial how appropriate candidates for each targeted therapy are selected, possibly using predictive biomarkers for drug response.

Declaration of interest

H Miyamoto has received research funding from Astellas Scientific and Medical Affairs, Ferring Research Institute, and Bristol-Myers Squibb. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

A peer reviewer has disclosed that they are currently Chair of the ANZUP BUP subcommittee, have served as remunerated Chair of recent Expert Opinion Forum for MSD, have received travel fees from AstraZeneca, Abbott, Astellas, and Novartis, have served as a remunerated advisory board member for Mundipharma and have attended remunerated speaker meetings for GSK and Novartis.

Additional information

Funding

This paper received no funding.