Leaders Establish Consensus Criteria for High-Risk Bladder Cancer Trials, Journal Club - Rashid Sayyid & Zachary Klaassen

December 21, 2023

Rashid Sayyid and Zach Klaassen discuss the Society for Immunotherapy of Cancer and the International Bladder Cancer Group's recommendations for clinical trial design in high-risk non-muscle-invasive bladder cancer. The consensus statement, led by Dr. Ashish Kamat and published in the Journal of Clinical Oncology, aims to guide investigators in late-phase clinical trial design. The panel, comprising multidisciplinary experts and a patient representative, used a modified Delphi process to develop recommendations based on literature evidence and clinical experience. They define high-risk non-muscle-invasive bladder cancer, including all high-grade tumors and carcinoma in situ, and categorize patients based on prior BCG exposure. The panel recommends specific study objectives, such as augmenting antitumor immune response and inducing durable complete responses. They outline detailed study design considerations, including eligibility criteria, control arms, primary and secondary endpoints, and evaluation methods.


Rashid Sayyid, MD, MSc, Urologic Oncology Fellow, Division of Urology, University of Toronto, Toronto, Ontario

Zachary Klaassen, MD, MSc, Urologic Oncologist, Assistant Professor Surgery/Urology at the Medical College of Georgia at Augusta University, Well Star MCG, Georgia Cancer Center, Augusta, GA

Read the Full Video Transcript

Rashid Sayyid: Hello, everyone, and thank you for joining us in this special UroToday Journal Club recording. I'm Rashid Sayyid, a urologic oncology fellow at the University of Toronto. And along with Zach Klaassen, Associate Professor and Program Director at Wellstar MCG Health, we'll be discussing the recommendations from the Society for Immunotherapy of Cancer and the International Bladder Cancer Group, specifically discussing the definitions, endpoints, and clinical trial design recommendations for bladder cancer. This is a four-part series Journal Club recording, and in this session, we'll be discussing the recommendations for the high-risk non-muscle-invasive bladder cancer subgroup. This consensus statement was published in the Journal of Clinical Oncology with Dr. Ashish Kamat as the first lead author.

By way of an introduction, as many of the urologic oncology community are aware, and many of our readers in particular, there's a significant unmet need for new and effective treatments for urothelial carcinoma. And so, in order to move this field forward, clinical trial design and conduct is going to be of utmost importance. And we need rationally selected endpoints, we discuss and agree on eligibility criteria, discuss evaluations, trial design, and statistical analysis, particularly with regards to powering for differences in the selected endpoints, and also discussing correlative studies to push the field forward.

And so, the objective of this consensus statement was to provide guidance to the investigators for late-phase clinical trial design from both the International Bladder Cancer Group and the Society for Immunotherapy of Cancer. This panel consisted of 25 members with multidisciplinary experience in clinical trial design, urology, surgery, medical oncology, radiation oncology, pathology, statistics, and patient advocacy. Importantly, this panel also consisted of member representatives from both the Society for Immunotherapy for Cancer and the International Bladder Cancer Group. Additionally, it included a patient representative from the Bladder Cancer Advocacy Network, which is critical in order to provide a voice and ensure that patients are heard when considering clinical trial design and conduct, particularly when they are the subject of such trials.

In terms of conflict of interest, this was very well regulated in this consensus statement, and the conflicts of interests were individually reported prior to the onset of manuscript development. No members with industry employment were eligible to serve on the panel, and the financial support for the development of this guideline was provided solely by the Society for Immunotherapy for Cancer and the IBCG.

In terms of recommendation development, the recommendations were based on literature evidence and clinical experience, where appropriate. And panel members participated in a modified Delphi process, in which an initial anonymous survey was administered to identify recommended endpoints. And subsequently, when practice-changing data or clinical trial results became available, authors considered and incorporated new evidence into the recommendations. And then these recommendations were refined through consensus discussions and reviewing and editing of manuscript drafts.

Now let's talk about the definition of high-risk non-muscle-invasive bladder cancer as defined by this panel. This included all high-grade tumors, Ta and T1, irrespective of the size. And the astute listener may note that this is different, or in contrast, to the AUA definition, which includes high-grade Ta, less than or equal to three centimeters, among the intermediate-risk non-muscle-invasive. Whereas this consensus panel recommends that any high-grade tumor, irrespective of the size, should be among the high-risk subgroup. And this also includes all patients with carcinoma in situ. They do also note the stratification into subgroups with variable prognostic features should be considered, such as the presence of CIS versus not, and multifocal high-grade T1 versus a small, unifocal, high-grade Ta. So, this is very important in stratifying this risk subgroup.

Now, further subclassification by prior exposure to BCG therapy is important and is unique to this cohort for the most part. BCG-naive patients have no prior BCG exposure. In terms of BCG-exposed, so these patients have received BCG within the prior 24 months, but not meeting the criteria for BCG-unresponsive. And then we have the third group of BCG-unresponsive, which includes patients with high-grade Ta or T1 disease recurring within six months, or persistent recurrent CIS with or without papillary disease within 12 months of adequate BCG, and/or high-grade T1 at the first evaluation following induction BCG. And what is adequate BCG? It's defined as at least five of six doses of induction BCG and at least two of three doses of maintenance BCG, or a second induction BCG in patients with recurrent persistent high-grade Ta or CIS, and five of six doses induction BCG in recurrent persistent T1 tumors.

Now, what are some pathology considerations in this risk subgroup? It's important to note that high-risk tumors are a heterogeneous subgroup, and so they include different variations. CIS, non-invasive Ta, and T1 tumors, which invade the lamina propria, are all included in this mixed group. It's also important to note that CIS is often multifocal and so can also exist in the upper urinary tract, resulting in positive high-grade cytology in a normal-looking bladder. And then, CIS can also present concurrent with high-grade Ta and high-grade T1 tumors. It's very important to be well versed in these nuances. It's also important to note that cytology can identify high-grade urothelial carcinoma, but does not provide any information on tumor stage. And, specifically within the T1 tumors, all of which are high-risk, assessment for histologic subtypes or adverse pathologic features, such as lymphovascular invasion, are important to prognosticate treatment response.

Now, what are the research hypotheses in this high-risk non-muscle-invasive bladder cancer subgroup? Further augmentation of the antitumor immune response for patients with high-risk non-muscle-invasive bladder cancer can include augmenting this with an investigational agent in addition to BCG, so we're furthering and adding an additional agent. A new strain of BCG, an alternative to BCG in both BCG-naive and BCG-exposed non-muscle-invasive bladder cancer patients, and after BCG, specifically in those with BCG-unresponsive non-muscle-invasive bladder cancer.

And the panel argued that this augmentation of the antitumor immune response may allow for more effective eradication of cancer cells compared with treatment with BCG alone, be it in BCG-naive or BCG-exposed, or compared to historical controls as well. So, it's very important that we further the response compared to what we see with BCG historically. And then, specifically in the BCG-unresponsive CIS, a randomized trial is recommended for papillary disease, thereby inducing a durable complete response for CIS, or a complete response versus prolonging event-free survival for Ta and T1 tumors. So, we need to think about it differently, given that with CIS, the goal is complete response in this setting.

And at this point, I'll turn it over to Zach, who will go over the study objectives in the high-risk non-muscle-invasive bladder cancer subgroup, and the rest of the study design considerations in this cohort.

Zach Klaassen:
Thanks so much, Rashid, for that excellent introduction. Let's talk about study objectives for high-risk disease. So, appropriate primary endpoints for patients with CIS, plus or minus papillary disease, are CR rate at six months, whereas for papillary-only disease, it is recurrence-free survival. In single-arm studies, the primary endpoint of CR rate at six months and a secondary endpoint of duration of CR are recommended in trials evaluating novel agents in BCG-unresponsive CIS. With regard to papillary-only disease, tumors are required to be fully resected prior to study entry, and treatment is in the adjuvant setting to prevent recurrence.

For statistical considerations. In randomized phase II trials, RFS is the primary endpoint, and deaths are included as an event in recurrence-free survival in high-risk non-muscle-invasive trials, whereas low-grade recurrence is not. For single-arm studies, enrolling patients with BCG-unresponsive CIS plus or minus papillary disease, CR rate at six months is the primary endpoint, and a 50% CR rate at six months is clinically meaningful, but the caveat is that this is a high bar that has only recently been attained in some studies. Finally, duration of response in complete responders should be a secondary endpoint and should be calculated using the Kaplan-Meier technique.

Looking at additional trial design considerations, when we look at the critical value for effect size or response rate threshold, this is broken down by BCG status. In patients with BCG-naive CIS, we should be looking for a CR rate of 70%. For BCG-exposed CIS, a CR rate of 60%. For BCG-unresponsive CIS, a CR rate of 50%. And for patients with BCG-naive and BCG-exposed Ta/T1, we should be assessing for a 10% increase in two-year RFS rate. And finally, for BCG-unresponsive Ta/T1, a one-year RFS rate of 30% should be our goal. For sample size calculations, assuming a two-year RFS of 70% in the control arm, there needs to be 150 events in the experimental arm for a hazard ratio of 0.69, an alpha of 0.05, and a beta of 0.20.

Looking at study design in more detail, for untreated BCG-naive high-risk non-muscle-invasive bladder cancer, the randomization to the investigational agent versus the control should be standard protocol, and control protocol for these patients is standard BCG induction followed by SWOG protocol maintenance. For those patients in the BCG-exposed setting, comparison of the investigational agent with a control arm receiving placebo plus BCG should be incorporated if the treatment arm includes BCG plus an investigational agent. Furthermore, in the BCG-unresponsive setting, random assignment is less acceptable, given that the FDA has allowed single-arm trials to be performed. For CIS, the CR rate and duration are clinically meaningful endpoints and should be benchmarked against historical values. And for BCG-unresponsive CIS, who have CIS recurrence at three months, one additional course of the investigational treatment is allowable. Finally, stratification criteria should include key prognostic factors.

For the study population, historically, prostatic urethral involvement has been excluded from trials, but the panel recommends that these should be included and stratified for random assignment. The panel recommended randomized studies with pooling of mixed populations of patients with CIS, with or without papillary disease, and papillary-only disease, where a composite endpoint of event-free survival can be used to evaluate the treatment effect of a novel therapy. And finally, histology subtypes should be allowed, and this has previously been an area of some contention, with a stratification factor for the presence of greater than 50% of the histological variation. This should be considered when including these patients in these trials.

With regards to evaluation and follow-up of high-risk non-muscle-invasive cancer, cystoscopy and urine cytology at three-month intervals and CT or MRI urography at six to 12-month intervals should be used for efficacy evaluation. If patients are undergoing advanced cystoscopic techniques such as blue light cystoscopy, this should be consistently used at baseline and on-treatment for these patients. There should be a recommended random 12-month bladder biopsy as an option for high-risk non-muscle-invasive bladder cancer trials.

Because the majority of recurrence or progression events will occur within the first two years from the start of treatment, these trials should be of adequate duration of at least two years, which also allows for adequate evaluation of key response endpoints, and also, we can follow these patients longer for the subsequent survival endpoints.

So, by way of summary, we'll quickly highlight again what we've just discussed in this nice tabular format. Eligibility for BCG-naive patients includes high-grade Ta and T1 tumors and CIS. For BCG-exposed, this includes high-grade papillary tumors or CIS, which do not meet the criteria for BCG-unresponsive disease. For BCG-unresponsive disease, Rashid already laid out this excellent definition, and you can read it here in the table on the left. With regards to the study design, this includes BCG-naive and BCG-exposed. This should be randomized controlled trials. However, in BCG-unresponsive patients, single-arm studies are allowable.

The control arm for BCG-naive patients should be BCG. For BCG-exposed, it may include additional BCG. And the primary endpoint for BCG-naive, exposed, and unresponsive CIS, plus or minus Ta/T1 disease, is a CR rate at three and/or six months, and for Ta/T1 only, it is recurrence-free survival. For secondary endpoints for both BCG-naive and BCG-exposed, you can see them listed here, highlighting specifically for the naive and exposed duration of CR and toxicity, and highlighting for BCG-unresponsive specifically event-free survival. And the primary endpoint assessment should be cystoscopy and urine cytology at three-month intervals and CT or MRI urography at six to 12-month intervals.

We thank you very much for your attention. We hope you enjoyed this UroToday discussion of the IBCG and the SITC recommendations for clinical trial design of high-risk non-muscle-invasive bladder cancer.