All five of the anti-PD-(L)1 agents for urothelial carcinoma are currently approved by the FDA as treatment for LA/mUC patients who have disease progression during or following platinum-based chemotherapy or within 12 months of neoadjuvant or adjuvant treatment for localized disease with platinum-based chemotherapy. In the trials leading to their approval ^{4– 8} , ORRs across all patients ranged from 15% with atezolizumab in IMvigor210 ⁶ to 21.1% with pembrolizumab in KEYNOTE-045 ⁸ . Median OS ranged from 6.5 months with avelumab ⁴ to 18.2 months with durvalumab ⁵ . Importantly, the phase III KEYNOTE-045 trial of pembrolizumab now has over 2 years of patient follow-up demonstrating a continued OS benefit over second-line chemotherapy with median OS of 10.1 months with pembrolizumab and 7.3 months with chemotherapy (hazard ratio [HR] 0.70, 95% confidence interval [CI] 0.57–0.85) ⁹ . The IMvigor211 study was a similarly designed phase III randomized trial comparing atezolizumab versus chemotherapy. However, the primary endpoint, OS, was tested hierarchically in prespecified populations—that is, IC2/3 (PD-L1 expression on at least 5% of tumor-infiltrating immune cells), followed by IC1/2/3, and then the intention-to-treat population. In the IC2/3 population, there was no significant difference in median OS (atezolizumab 11.1 months versus chemotherapy 10.6 months; HR 0.87, 95% CI 0.63–1.21), precluding further formal statistical analyses in the other prespecified populations and thereby resulting in an overall negative study ¹⁰ .

Investigation into the use of immunotherapy combinations post-platinum is ongoing, but early data are promising ( Table 1). The phase II CheckMate 032 trial compared nivolumab monotherapy with nivolumab 3 mg/kg plus ipilimumab 1 mg/kg (NIVO3+IPI1) and nivolumab 1 mg/kg plus ipilimumab 3 mg/kg (NIVO1+IPI3) ¹¹ . In PD-L1 unselected patients, ORR ranged from 25.6% with nivolumab alone to 38% with NIVO1+IPI3. In patients with PD-L1 expression of at least 1%, ORRs were 26.9% with nivolumab alone but 58.1% with NIVO1+IPI3. Furthermore, median OS was 9.9 months with nivolumab alone and 15.3 months with NIVO1+IPI3 across all patients but was 12.9 and 24.1 months, respectively, in patients with PD-L1 expression of at least 1%. Although grade 3 or 4 treatment-related adverse events were more common with NIVO1+IPI3 compared with nivolumab (39.1% versus 26.9%), these results suggest that combination therapy may provide a significant benefit over monotherapy, particularly for patients whose tumors express PD-L1.

There is also a newly established role for anti-PD-(L)1 agents as switch maintenance therapy following completion of first-line platinum-based chemotherapy. The phase III JAVELIN Bladder 100 trial randomly assigned 700 LA/mUC patients whose disease did not progress after first-line platinum-based chemotherapy to receive maintenance avelumab plus best supportive care versus best supportive care alone. At the planned interim analysis, patients receiving maintenance avelumab had a significant improvement in median OS over best supportive care alone (21.4 versus 14.3 months; HR 0.69, 95% CI 0.56–0.86) as well as a progression-free survival (PFS) benefit (3.7 versus 2.0 months; HR 0.62) ¹² . These results led to the recent FDA approval of avelumab switch maintenance therapy following first-line chemotherapy in patients with mUC. Similarly, the phase II HCRN GU14-182 study enrolled LA/mUC patients who achieved at least stable disease following first-line platinum-based chemotherapy and randomly assigned them to receive maintenance pembrolizumab versus placebo ¹³ . Patients receiving maintenance pembrolizumab demonstrated an improvement in median PFS compared with placebo (5.4 versus 3.0 months; HR 0.65).

In addition to approval for patients who progress following platinum-based chemotherapy, atezolizumab and pembrolizumab are approved in the first-line setting for LA/mUC. Both agents were initially approved as first-line treatment for cisplatin-ineligible patients on the basis of the phase II IMvigor210 ¹⁴ and KEYNOTE-052 ¹⁵ trials. Subsequently, the randomized phase III IMvigor130 trial of atezolizumab and KEYNOTE-361 trial of pembrolizumab enrolled platinum-eligible patients with LA/mUC and no prior systemic therapy to receive atezolizumab/pembrolizumab with or without platinum-based chemotherapy versus platinum-based chemotherapy alone ¹⁶ . In June 2018, interim analyses of these two trials showed that patients with low PD-L1 expression receiving atezolizumab or pembrolizumab monotherapy had decreased survival compared with patients with low PD-L1 expression who received platinum-based chemotherapy, leading to a change in drug approval ¹⁶ . Currently, atezolizumab and pembrolizumab are indicated as first-line treatment for LA/mUC patients who are cisplatin-ineligible and whose tumors express PD-L1 or patients who are not eligible for any platinum therapy regardless of PD-L1 status. It is important to note that the final analysis of either trial has not been published, but results from IMvigor130 have been presented. Similarly, the phase III DANUBE trial compared durvalumab monotherapy or durvalumab plus tremelimumab versus platinum-based chemotherapy. A press release stated that the study did not meet its primary endpoints for OS in high–PD-L1 patients who received durvalumab or OS in patients who received durvalumab plus tremelimumab regardless of PD-L1 status, but results have not yet been presented or published ¹⁷ .

There is also growing evidence to suggest that the clinical benefit of the combination of immune checkpoint inhibitors with chemotherapy in the first-line setting may be limited. Results from IMvigor130 evaluating the combination of atezolizumab plus chemotherapy demonstrated a PFS benefit over chemotherapy alone (8.2 versus 6.3 months; HR 0.82, 95% CI 0.70–0.96) but this benefit was small and of questionable clinical significance ¹⁸ . One concern regarding the outcome of IMvigor130 is that 40% of patients deemed cisplatin-eligible received carboplatin-based chemotherapy, yet subgroup analysis suggested an overall survival benefit seen only in the cisplatin-treated patients. A recent press release also reported that, in KEYNOTE-361, pembrolizumab plus chemotherapy did not meet the dual primary endpoints for superiority in OS or PFS over chemotherapy alone ¹⁹ . Similar trials are ongoing to further evaluate first-line immunotherapy plus chemotherapy, including CheckMate901 comparing first-line nivolumab plus ipilimumab or standard-of-care chemotherapy versus standard chemotherapy (NCT03036098) and NILE comparing durvalumab plus chemotherapy with or without tremelimumab with chemotherapy alone (NCT03682068) ( Table 2).

Preferred management of MIBC includes neoadjuvant cisplatin-based chemotherapy prior to radical cystectomy (RC), and pathologic complete response (pCR) at cystectomy is associated with increased OS ²⁰ . Immunotherapy has not yet been approved in the neoadjuvant setting for MIBC, but some preliminary studies show promise ( Table 3). The phase II PURE-01 study enrolled 50 patients with clinical T2-3bN0M0 MIBC and administered three doses of pembrolizumab prior to RC ²¹ . At cystectomy, 42% of patients had a pCR and 54% of patients had pathologic downstaging to less than pT2. Among patients with a PD-L1 combined positive score (CPS) of at least 10%, 54.3% achieved a pCR and 65.7% were downstaged to less than pT2 whereas only 13.3% and 26.7% of patients with PD-LI of less than 10% achieved these same outcomes. A significant association between tumor mutation burden (TMB) and pCR was also seen. The similar phase II ABACUS trial administered two cycles of neoadjuvant atezolizumab and observed a pCR rate of 31% ²² . In contrast to PURE-01, the ABACUS trial found no significant correlation between PD-L1 expression or TMB with pCR or 1-year relapse-free survival rates, but patients with high intraepithelial CD8 ⁺ cells had a significantly higher pCR rate compared with those without CD8 ⁺ cells (40% versus 20%, P <0.05). These conflicting biomarker results suggest that additional research is needed to clarify the best biomarker for predicting response to immunotherapy in bladder cancer.

Clinical trials have also combined neoadjuvant immune checkpoint inhibitors plus chemotherapy in MIBC. Recent results from the BLASST-1 trial of neoadjuvant nivolumab with gemcitabine and cisplatin demonstrated a pCR rate of 49% and downstaging to less than pT2 in 65.8% of patients ²³ . Similarly, results of a phase Ib/II trial of neoadjuvant pembrolizumab plus gemcitabine and cisplatin reported a pCR rate of 44%, and 61% of patients were downstaged to less than pT2 ²⁴ . With a median follow-up of 14 months, the estimated 12-month relapse-free survival was 80% and OS was 94%. Together, these studies suggest that immunotherapy will likely have a role in the management of patients with MIBC. Trials are under way to further evaluate perioperative immunotherapy as monotherapy, immunotherapy in rational combinations, and chemo-immunotherapy approaches.

For patients who are either not eligible for or not interested in RC, TMT is an alternative treatment option. TMT involves a maximal transurethral bladder tumor resection followed by concurrent chemoradiation. Several trials are investigating the addition of immunotherapy to chemoradiation in patients with MIBC in an attempt to harness the abscopal effect. KEYNOTE-992 is a phase III trial comparing pembrolizumab with chemoradiation versus placebo with chemoradiation, followed by pembrolizumab or placebo every 6 weeks for up to a year, SWOG/NRG 1806 is evaluating the use of chemoradiation with or without atezolizumab, and another trial is evaluating pembrolizumab and gemcitabine with concurrent radiation therapy (NCT02621151). Results from these trials are not yet available but are eagerly anticipated ( Table 4).

The first FDA approval for immunotherapy in the non-metastatic setting came in January 2020 when pembrolizumab was approved for patients with BCG-unresponsive, high-risk NMIBC with carcinoma in situ (CIS) with or without papillary tumors who are ineligible for or have chosen not to undergo cystectomy. This approval was based on results from cohort A of the KEYNOTE-057 trial showing a 3-month complete response (CR) rate of 40% with 46% of responses lasting at least 12 months ²⁵ . Similar results were seen in the phase II SWOG S1605 trial of atezolizumab in BCG-unresponsive NMIBC. Preliminary results from 73 patients showed a 3-month CR rate of 41% and 6-month CR rate of 26% ²⁷ . Additional trials evaluating the use of immunotherapy in NMIBC are ongoing; these include the phase III KEYNOTE-676 trial of BCG with or without pembrolizumab (NCT03711032), a phase II trial of gemcitabine plus pembrolizumab (NCT04164082), and the phase II ADAPT-Bladder trial comparing durvalumab monotherapy, durvalumab plus BCG, and durvalumab plus external beam radiation (NCT03317158).

Erdafitinib is a tyrosine kinase inhibitor of FGFR1–4 and the first targeted therapy approved for mUC. The phase II trial of erdafitinib included 99 patients whose tumor harbored an FGFR3 mutation or FGFR2/3 fusion and who had disease progression following chemotherapy ³³ . The confirmed ORR was 40% and an additional 39% of patients had stable disease. A total of 22 patients had previously received immunotherapy with only one achieving a response, yet the response rate for erdafitinib for this subgroup was 59%. At a median follow-up of 24 months, the median PFS was 5.5 months (95% CI 4.0–6.0) and the median OS was 11.3 months (95% CI 9.7–15.2) ³⁴ . Based on these results, erdafitinib was approved by the FDA in April 2019 for patients with mUC with a susceptible FGFR2/3 alteration following platinum-containing chemotherapy. Multiple ongoing trials—including a phase II study of erdafitinib alone or in combination with cetrelimab, an anti-PD-1 antibody, and as first-line therapy for cisplatin-ineligible patients with mUC and an FGFR alteration (NCT03473743)—are assessing erdafitinib in other clinical scenarios.

In addition to erdafitinib, other FGFR inhibitors are under investigation. An expansion cohort to the phase I trial of infigratinib included 67 mUC patients who progressed on or had contraindications to platinum-based chemotherapy and whose tumor harbored an alteration in FGFR3 ³⁵ . The confirmed ORR was 25.4%, median duration of response was 5.06 months, median PFS was 3.75 months, and median OS was 7.75 months. A subsequent analysis of this same cohort observed that patients with upper tract urothelial cancer (UTUC) had a confirmed ORR of 50% and a disease control rate (DCR) of 100% but that those with urothelial cancer of the bladder (UCB) had an ORR of 22% and a DCR of 59.3% ³⁶ . Additionally, the median PFS and median OS were 8.54 and 21.82 months for those with UTUC and 3.65 and 7.0 months for those with UCB. Prior work has shown that FGFR3 alterations are more common in UTUC than UCB (40% versus 26%) ³⁷ and thus UTUC may be more amenable to FGFR inhibition. Although this study included a small number of patients with UTUC, these initial results certainly warrant further evaluation and a phase I/II trial of neoadjuvant infigratinib for patients with UTUC is planned (NCT04228042). Additionally, a phase III trial comparing adjuvant infigratinib versus placebo in patients with FGFR3 alterations and high risk for disease recurrence is under way (NCT04197986).

Chromatin structure can be modified via many mechanisms, including histone acetylation/deacetylation and histone methylation/demethylation, resulting in regulation of gene transcription. Disruption of this process is implicated in the pathogenesis of urothelial cancer and therefore may be a viable target for new therapies, such as histone deacetylase (HDAC) inhibitors and enhancer of zeste homolog 2 (EZH2) inhibitors ³⁸ . Mocetinostat, a class I/IV HDAC inhibitor, was administered to 17 patients with mUC with progression after platinum-based chemotherapy and an inactivating mutation or deletion in CREBBP, EP300 ³⁹ , or both. The ORR was 11% in stage 1 and so the study was terminated. Although mocetinostat did not appear to be effective in this cohort of patients with mUC, it is possible that a different biomarker is needed to predict patient response.

Pre-clinical studies have demonstrated that EZH2 inhibition induces cell death in models of urothelial cancer ^{40, 41} . Additionally, the response appears to be enhanced when the EZH2 inhibitor tazemetostat is combined with an anti-PD-1 antibody ⁴² . Based on these findings, a phase I/II trial evaluating the combination of tazemetostat plus pembrolizumab in patients with either cisplatin-refractory or cisplatin-ineligible mUC is under way (NCT03854474). Similarly, EZH2 inhibition has been shown to improve the response to anti-CTLA-4 therapy in a murine model of bladder cancer ⁴³ . This led to the phase Ib/II ORIOn-E trial of the EZH2 inhibitor CPI-1205 combined with ipilimumab, which includes a cohort of patients with mUC (NCT03525795). This trial is currently closed to accrual, but results have not yet been released. Other agents targeting chromatin modification genes appear promising in pre-clinical studies and are anticipated to move into early-phase clinical trials in the near future.

Multiple agents targeting HER2, including trastuzumab, lapatinib, and afatinib, have also been tested in patients with urothelial cancer ⁴⁴ . Results thus far have been somewhat mixed, possibly partially owing to inclusion of HER2 unselected patients and the discordance in HER2 classification between immunohistochemistry (IHC), fluorescence in situ hybridization, and molecular characterization. A phase II study of afatinib in HER2 unselected patients with platinum-refractory mUC found an ORR of 8.6% and a median PFS of 1.4 months for the entire cohort ⁴⁵ . However, they also found that 83% (5/6) of patients with HER2 copy number amplification or ERBB3 somatic mutations (or both) achieved a PFS of at least 3 months but that 0% of patients without alterations did. Interestingly, no correlation between IHC for ERBB3, HER2, or EGFR and clinical response to afatinib was seen. Additional trials of HER2 targeted agents, including a follow-up study of afatinib in patients with alterations in EGFR, HER2, ERBB3, or ERBB4 (NCT02122172), are ongoing. It remains to be seen whether this will prove to be a viable treatment option in appropriately selected patients.

EV is an ADC targeting Nectin-4, a cell adhesion molecule highly expressed in nearly all urothelial tumors, conjugated to monomethyl auristatin E, a microtubule-disrupting agent ⁴⁷ . In the dose expansion portion of the EV-101 trial, 112 mUC patients who failed at least one prior therapy received EV ⁴⁸ . The confirmed ORR was 43%, including 5% CR, and the median duration of response was 7.4 months. Subsequently, EV-201 enrolled patients with mUC treated with prior platinum and anti-PD-(L)1 (cohort 1) or treated with prior anti-PD-(L)1 and cisplatin-ineligible (cohort 2) ⁴⁹ . Cohort 1 enrolled 125 patients with a confirmed ORR of 44%, including a 12% CR rate. Responses were seen across subgroups including an ORR of 41% in non-responders to prior anti-PD-(L)1 and 38% in patients with liver metastases. Results for cohort 2 have not yet been released.

EV-103 is an ongoing multi-cohort trial of EV alone or in combination with other therapies and includes cohorts of patients with mUC and localized MIBC. Cohort A evaluated EV plus pembrolizumab as first-line treatment for 45 cisplatin-ineligible patients with LA/mUC ⁵⁰ . The confirmed ORR was 73.3%, including 15.6% CRs, and the DCR was 93.3%. With a median follow-up of 10.4 months, the median duration of response was not yet reached and the median PFS was 12.3 months. These results are extremely encouraging, particularly for cisplatin-ineligible patients who have limited effective treatment options. Additional study cohorts, including EV plus chemotherapy as first-line treatment for mUC and EV alone or with pembrolizumab as neoadjuvant therapy for localized MIBC (NCT03288545), are ongoing.

Sacituzumab govitecan (SG) is an ADC employing an anti-Trop-2 antibody conjugated to SN-38, the active metabolite of irinotecan ⁵¹ . Trop-2 is a transmembrane glycoprotein that is important for cell growth and tumorigenesis and that is overexpressed in urothelial cancer. The phase I/II basket trial of SG included 45 mUC patients who had received at least one prior line of systemic therapy ⁵² . The ORR was 31%, median duration of response was 12.6 months, median PFS was 7.3 months, and median OS was 18.9 months. The subsequent phase II TROPHY-U-01 trial of SG recently completed accrual for cohort 1, which enrolled 100 mUC patients who progressed after platinum-based therapy and a checkpoint inhibitor ⁵³ . Preliminary results from 35 patients demonstrated an ORR of 29% and in light of these data the FDA granted SG fast-track designation for mUC in April 2020 ⁵⁴ . Accrual to two additional patient cohorts—a cohort of platinum-ineligible patients who progressed following checkpoint inhibitor therapy and a cohort of immune checkpoint inhibitor-naïve patients who will receive SG plus pembrolizumab—is ongoing.