Should chemotherapy still be used to treat all muscle invasive bladder cancer in the “era of immunotherapy”?

KEYWORDS:

• Muscle invasive bladder cancer
• chemotherapy
• immunotherapy

1. Introduction

Platinum-based chemotherapy has served as the mainstay systemic treatment for muscle invasive bladder cancer (MIBC) for decades. Its use in the neoadjuvant (NAC), adjuvant (AC), and salvage setting has led to improved overall survival for those with localized disease and prolonged life for those with metastatic disease [1,2]. Recently, we have come to appreciate that MIBC is a heterogeneous disease that does not uniformly respond to cisplatin-based chemotherapy. Furthermore, there are many patients with organ-confined MIBC who are appropriately managed by radical cystectomy alone [3]. In turn, this begs the question whether cisplatin-based chemotherapy is the appropriate mainstay systemic treatment for all patients with MIBC today.

Immunotherapy has taken the limelight as a systemic treatment option for a multitude of cancers. Many view it as less toxic than conventional chemotherapy and because of the complete responses seen in some patients; some believe it to be the future of systemic therapy. In bladder cancer alone we have witnessed the approval of multiple immune checkpoint inhibitors agents over the past two years.

As we move forward using inherent tumor and patient characteristics to provide ‘personalized medicine’, we must step-back and re-evaluate our current practice patterns. In doing so, we must assess the role of cisplatin-based chemotherapy in this new era of immuno-oncology.

2. Cisplatin-based chemotherapy for MIBC

Over the past decade, significant effort has been put forth in order to better understand the molecular biology of MIBC, with the goal of improving patient selection for perioperative systemic chemotherapy. By using gene expression profiling techniques, we have been able to classify MIBC’s into distinct molecular subtypes, which at the simplest level can be divided into basal and luminal subtypes [4]. The clinical implications of such a classification system have been depicted in several retrospective studies demonstrating that basal subtype tumors are more responsive to cisplatin-based chemotherapy than their luminal counterparts, which appear to possess a chemo-resistant phenotype [4,5].

In addition to gene expression profiling, mutational analysis, in particular analysis of mutations in DNA damage repair (DDR) genes (ERCC2, RB1, ATM, and FANCC) have shown prognostic value by selecting for patients who achieve optimal response (ideally pT0) with NAC [6]. Response to platinum-based chemotherapy in the metastatic setting also correlates with DDR mutations [7].

Although we have made significant strides forward in the molecular characterization of MIBC and identification of chemosensitive tumor subtypes, several shortcomings still exist making routine application of these discoveries not yet a reality in the day-to-day management of MIBC. As with many things in cancer medicine, the story is not simply black and white. Several factors such as tumor heterogeneity (both inter- and intra-tumoral), clonal evolution in response to therapy, as well as the practical aspects such as cost per patient and access to testing remain significant obstacles in the widespread implementation of these discoveries. Until we can overcome these aforementioned barriers and develop cost-efficient biomarkers to reliably identify chemosensitive tumors, we will continue to administer perioperative cisplatin-based chemotherapy guided by only clinical and pathologic features, and not biomarkers.

It is important we remember that cisplatin-based chemotherapy, although effective, is not a benign treatment. Significant oto, neuro, and renal toxicity are potential side effects associated with this therapy. For this reason, not all patients receive the prescribed dose and some are not eligible for the therapy at all given their medical co-morbidities (ie. ‘cisplatin-ineligible patients’). Given the fact that bladder cancer is typically a disease of the elderly, the number of ‘cisplatin-ineligible’ patients is not trivial [8]. Carboplatin-based regimens are less nephrotoxic, however, are not recommended [9].

Until we can accurately predict response to chemotherapy in an economically feasible way, I believe we will continue to use this proven therapeutic regimen despite its potential toxicity as it has stood the test of time in the treatment of MIBC.

3. Immunotherapy for MIBC

Urothelial cancer is renowned as a highly immunogenic tumor possessing a high mutational burden relative to other malignancies [10]. The use of immunotherapy for bladder cancer treatment is not a novel concept as BCG has been used in the non-muscle invasive setting for decades. Over the past several years, we have witnessed an eruption of immune checkpoint therapy utilization across a variety of malignancies. In several cancer types, we have seen impressive and durable responses, with a subset of patients being ‘cured’ with complete responses. In bladder cancer, five agents have been approved since 2016 in the setting of progression after cisplatin-based chemotherapy and two agents approved in the NAC setting for cisplatin-ineligible patients [10].

Despite the success of immune checkpoint inhibition in MIBC, we must remember that similar to the case with chemotherapy not all patients respond equally. In fact, only 13–21% of patients with advanced urothelial cell cancer respond to immune checkpoint inhibition [11]. Preliminary research using the molecular subtyping schema described above indicate that response to immunotherapy may be predicted with this subtyping system as well. Two clinical trials studying immune-checkpoint blockade in metastatic MIBC found that the luminal-infiltrated subtype, a subtype characterized by high levels of immune cell infiltration, appears the most responsive to treatment with immune checkpoint inhibition [12,13]. In addition to molecular subtyping, studies investigating other biomarkers such as: total mutational burden, DDR mutations, and the importance T cell clonality are ongoing in hopes of identifying a potential marker(s) of response to immune checkpoint blockade [11]. We have also come to realize that immune gene signatures may aid in predicting response to immunotherapy. For example, in the Checkmate 275 study, a 25-gene interferon gamma signature was associated with improved response to immune checkpoint inhibition [13]. In the future, immune gene signatures may aid in predicting response to immune checkpoint inhibition by revealing information about the tumor microenvironment, allowing us to describe tumors as ‘hot’ and ‘cold.' However, until we have reliable, accurate, and cost-efficient biomarkers to predict responders, we will not be optimizing the use of these novel treatments. A better understanding surrounding the lack of response and patterns of resistance to immune checkpoint inhibition will help in the development of these much-needed biomarkers.

Although not directly cytotoxic like chemotherapy, immunotherapy is associated with its own unique and distinct adverse events, some of which can be life threatening. Specifically, in the treatment of metastatic bladder cancer immune-related adverse events can occur in up to 19% of patients (all grades) [14]. For this reason, we must continue our current research efforts to identify responders and explore mechanisms of resistance to avoid people suffering the ill effects of such a therapy with minimal therapeutic efficacy.

4. Final thoughts

As our therapeutic options for MIBC continue to grow, we as clinicians are tasked with choosing the appropriate strategy given individual patient characteristics. As is often the case with new therapies, we are drawn to their novelty, sometimes losing sight of the benefit our mainstay treatments have provided for decades. There is no doubt that immune checkpoint inhibition for MIBC is effective in a subset of patients, and for this reason, is here to stay. However, in our opinion, it has/will not eliminate the need for conventional cisplatin-based chemotherapy in this disease. As eluded to several times in this article, we must continue our research efforts to better understand at a molecular level the sensitivity and resistance patterns of MIBC to both chemotherapy and immune checkpoint blockade to optimize their use. Currently, validated and trustworthy biomarkers of response are lacking for the utilization of immune checkpoint inhibition in MIBC. The tissue-based biomarker ‘PDL1 expression’ has been used in clinical trials, however,due to lack of standardization and flaws with methodology its utility has been called into question [11]. The heterogeneous nature of this malignancy and lack of effective biomarkers cause significant challenges in choosing the ‘perfect’ patient for treatment with immune checkpoint blockade. Until such biomarkers are refined, systemic treatment with cisplatin-based chemotherapy will remain integral in managing MIBC.

Furthermore, we must explore the optimal sequencing of cisplatin-based chemotherapy and immune checkpoint blockade for long-term success in this disease. Importantly, these research efforts should not be unbalanced in favor of the ‘new kid on the bloc’ (i.e. immune checkpoint inhibition) because cisplatin-based chemotherapy still plays an integral role in the management of MIBC in 2019.

Declaration of interest

CP Dinney declares receiving consulting fees from FKD Therapies Oy and Janssen, consulting and research fees from Merck and the National Cancer Institute and research fees from the University of Eastern Finland. 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

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.