ICIs in Bladder Cancer
My Thoughts on New and Exciting Advances with Immune Checkpoint Blockade for Treatment of Bladder Cancer

Released: November 19, 2020

Expiration: November 18, 2021

Matthew Galsky
Matthew Galsky, MD

Activity

Progress
1
Course Completed

Immune checkpoint blockade for the treatment of bladder cancer evolved similarly to other systemic therapies in oncology. These agents were first tested in later stages of disease and moved forward once shown to be potentially effective and safe. Between 2015 and 2018, checkpoint blockade in bladder cancer was used in 2 major settings: (1) as a second‑line treatment for patients with metastatic disease who progressed following platinum‑based chemotherapy and (2) as a first‑line treatment for patients with metastatic cancer who were cisplatin ineligible. Between 2018 and 2020, this treatment paradigm changed.

In 2018, the data safety monitoring committees from the IMvigor130 and KEYNOTE-361 trials reported independently that either single-agent pembrolizumab or atezolizumab reduced early OS rates in patients with PD-L1–low tumors. Therefore, the FDA and the European Medicines Agency restricted first-line use of pembrolizumab and atezolizumab to cisplatin-ineligible patients whose tumors were PD-L1 high.

In early 2020, pembrolizumab gained approval in the United States for patients with BCG-unresponsive high-risk, non–muscle invasive bladder cancer, specifically carcinoma in situ. This approval marks the earliest clinical disease state for which an immune checkpoint inhibitor has been approved in any tumor type.

Finally, in June 2020, the FDA approved avelumab as a maintenance therapy for patients with locally advanced or metastatic urothelial carcinoma whose disease has not progressed after 4-6 cycles of first-line platinum-containing chemotherapy. We will briefly discuss this new approval and some additional updates on the use of immune checkpoint inhibitors for patients with bladder cancer.

Recent Trial Data
In 2020, results were reported from 4 large, randomized phase III clinical trials in metastatic urothelial cancer which asked a series of questions, as follows:

Should chemotherapy be given with immune checkpoint blockade as a first-line treatment for bladder cancer?
As mentioned above, the phase III IMvigor130 and KEYNOTE‑361 trials addressed the use of immune checkpoint inhibitors and chemotherapy as first-line treatments for advanced disease. The studies had similar designs and coprimary endpoints.

IMvigor130 was a phase III trial exploring the use of atezolizumab plus platinum-based chemotherapy, atezolizumab alone, or chemotherapy alone with co-primary endpoints of PFS and OS with atezolizumab plus platinum-based chemotherapy vs platinum-based chemotherapy alone, and OS with atezolizumab vs platinum-based chemotherapy. Median PFS was improved with atezolizumab plus platinum-based chemotherapy vs platinum-based chemotherapy alone (HR: 0.82; P = .007). In the interim OS analysis, there was an improvement in OS with atezolizumab plus platinum-based chemotherapy vs platinum-based chemotherapy alone, but this did not meet the prespecified threshold for significance at the interim analysis (HR: 0.82; P = .027) and OS will be further assessed at the final analysis. Notably, due to the hierarchical statistical analysis plan, the co-primary end point of OS with atezolizumab vs platinum-based chemotherapy was not formally compared; however, despite the early portions of the OS curve favoring platinum-based chemotherapy the and there was no appreciable difference in OS with atezolizumab vs platinum-based chemotherapy (HR: 1.02).

The phase III KEYNOTE‑361 trial assessed pembrolizumab plus platinum-based chemotherapy vs pembrolizumab alone vs chemotherapy alone with coprimary endpoints of PFS and OS for pembrolizumab plus platinum-based chemotherapy vs chemotherapy alone.  This trial did not meet the statistical thresholds for either of the coprimary endpoints (HR for PFS: 0.78; P = .003; HR for OS: 0.86; P = .0407).

Currently, based on the available results of both IMvigor130 and Keynote-361, concurrent platinum-based chemotherapy plus PD-1/PD-L1 blockade has not been adopted as a standard treatment but the final results of these trials are still awaited.

If patients with bladder cancer have at least stable disease after initial platinum-based chemotherapy, should an immune checkpoint inhibitor be used as a switch maintenance treatment?
Switch maintenance is when patients who have at least stable disease following initial therapy are switched to a second class of therapy as maintenance instead of the historical practice of stopping treatment and observing until progression. Two randomized trials in bladder cancer, HCRN GU14‑182 and JAVELIN Bladder 100, evaluated switch maintenance with immune checkpoint inhibitors after platinum-based chemotherapy.

HCRN GU14‑182 was a phase II study that randomized patients to switch maintenance with pembrolizumab or placebo after at least SD with ≤ 8 cycles of platinum-based chemotherapy. A significant improvement in PFS, the primary study endpoint, was observed (HR: 0.64; P = .038). There was no significant improvement in OS, a secondary endpoint, though the study was not adequately powered to definitively assess this endpoint.

JAVELIN Bladder 100 was a large phase III trial that randomized patients whose disease had not progressed following 4-6 cycles of platinum-based chemotherapy to switch maintenance avelumab or observation. Significant PFS and OS benefits were observed with maintenance avelumab (HR for PFS: 0.62; P < .001; HR for OS: 0.69; P < .001), leading avelumab switch maintenance to be approved in the United States. This has quickly become a standard treatment strategy because of the considerable effect size in terms of the improvement in OS that was observed in this study.

Does combination immune checkpoint blockade provide a survival benefit as a first-line treatment in urothelial cancer?
The DANUBE trial, which was reported at 2020 ESMO, tested the combination of durvalumab and tremelimumab (an anti–CTLA-4 antibody) vs durvalumab alone vs chemotherapy in patients with previously untreated, unresectable, locally advanced or metastatic urothelial cancer. There was no significant improvement in OS with durvalumab plus tremelimumab vs platinum-based chemotherapy in the all comer population, a co-primary endpoint of the study. However, in the PD-L1 high population, an improvement in OS was observed with durvalumab plus tremelimumab vs platinum-based chemotherapy. As this was a secondary endpoint, there is not yet a standard role for therapy with an anti–PD-L1/PD-1 and anti–CTLA-4 antibody combination in patients with advanced bladder cancer. However, the role of ipilimumab plus nivolumab vs platinum-based chemotherapy in patients with PD-L1 high expressing tumors is being explored as a co-primary endpoint in the ongoing phase III CheckMate901 study (NCT03036098).

What is the therapeutic potential for immune checkpoint blockade without chemotherapy in patients with PD-L1–high tumors?
IMvigor130, KEYNOTE‑361, and DANUBE all addressed whether or not immune checkpoint blockade without chemotherapy was a first-line treatment strategy that should be embraced in all patients. Looking at the survival curves from these studies, they initially favor chemotherapy. Later on, immune checkpoint blockade shows a slight improvement vs chemotherapy, but this is not statistically significant. As a result, immune checkpoint blockade is not used as a first‑line treatment in unselected patients.

Remember, the FDA included the label restrictions for atezolizumab and pembrolizumab to patients with locally advanced or metastatic urothelial carcinoma who are cisplatin-ineligible and have high PD-L1 expression due to the preliminary data from the IMvigor130 and KEYNOTE‑361 trials. Comparisons of single-agent PD-1/PD-L1 blockade with platinum-based chemotherapy has been performed in the IMvigor130, KEYNOTE‑361, and DANUBE trials in both the all comer population and the subset of patients with PD-L1–high expressing tumors. The survival curves for the all-comer populations from these 3 studies were quite similar, with the initial portions of the curve favoring platinum-based chemotherapy and then the curves crossing at the 9-12–month timeframe. The OS analysis in patients with PD-L1–high expressing tumors are much more difficult to compare across the trials as each trial used a different assay for PD-L1 expression leading to identification of different subsets of patients, so I will address them individually.

In the IMvigor130 study, 24% of patients enrolled had high PD-L1 expression (IC2/3). A subgroup analysis suggested that OS was better in PD-L1–high patients treated with atezolizumab vs chemotherapy; however, because of the way the statistical analysis plan was structured, this was not a formal analysis and was an interim look at the data. Nevertheless, it does suggest a benefit for immune checkpoint blockade in the select group of patients with PD-L1–high expression based on the SP142 assay used in this trial.

In the KEYNOTE-361 study, approximately 45% of patients had high PD-L1 expression (combined positive score CPS ≥ 10). A subgroup analysis of patients with high PD-L1 expression showed no significant difference in OS with pembrolizumab monotherapy vs chemotherapy. In the DANUBE trial, 60% of patients had high PD-L1 expression and there was no improvement in outcomes with durvalumab vs platinum-based chemotherapy.

Together, these disparate results with different PD-L1 expression assays for the different PD-1/PD-L1 inhibitors complicate an understanding of the role of single-agent PD-1/PD-L1 blockade as first-line treatment in patients with metastatic urothelial cancer. However, importantly, the analyses presented to date have compared PD-1/PD-L1 blockade with platinum-based (cisplatin- or carboplatin-based) chemotherapy and have not been restricted to the cisplatin-ineligible subset of patients, the current labeled indication for pembrolizumab and atezolizumab.

Should we be moving immunotherapy to earlier in the disease course for patients with urothelial carcinoma?
Numerous large, international, randomized studies are testing immune checkpoint blockade alone or in combination with chemotherapy or other agents, like antibody–drug conjugates, as neoadjuvant or adjuvant treatment for patients with urothelial carcinoma. Recently, we have seen early data from 2 randomized trials in the adjuvant setting.

The phase III IMvigor010 trial randomized patients with high risk muscle invasive bladder cancer or cancer of the upper urinary tract to receive adjuvant atezolizumab or observation. Patients enrolled on this trial had undergone surgery and included 2 groups: (1) patients who did not receive neoadjuvant chemotherapy (either cisplatin-ineligible or declined) and had pT3-pT4a or pN+ tumors; and (2) patients who had previously received neoadjuvant chemotherapy and had ypT2-T4a or ypN+ tumors. IMvigor010 did not show a disease-free survival improvement in either the all‑comer or PD-L1–high patient populations with atezolizumab.

The phase III CheckMate 274 study (NCT02632409) of adjuvant nivolumab was designed similarly to IMvigor010, but included a placebo control, in patients with high-risk muscle invasive urothelial cancer after surgery. Similar to IMvigor010, enrolled patients could have received neoadjuvant chemotherapy or not prior to resection. According to a September 2020 press release, all‑comer and PD-L1–high patients who received nivolumab had significantly improved disease-free survival vs those who received placebo. If effective, nivolumab would be the first potential treatment for cisplatin-ineligible patients in this setting and for patients with residual disease after neoadjuvant cisplatin‑based chemotherapy. It would also be the first standard use of an immune checkpoint inhibitor for the perioperative treatment of urothelial cancer.

What challenges have you encountered in using immune checkpoint inhibitor therapy in your patients with bladder cancer? Share your thoughts in the comment box below!