There are substantial unmet needs for patients with metastatic urothelial carcinoma (UC). First-line cisplatin-based chemotherapy regimens yield a median survival of 12 to 15 months and long-term survival in 5% to 15%. Salvage systemic therapy yields a median survival of 6 to 8 months. Hence, the discovery of novel therapeutic targets is of paramount importance. Recent molecular analyses have provided insights regarding molecular tumor tissue alterations on multiple platforms. A multidisciplinary effort using innovative clinical trial designs and exploiting preclinical signals of robust activity guided by predictive biomarkers may provide much needed clinical advances in therapy for advanced UC.
Key points
- •
Current systemic therapy for metastatic urothelial carcinoma yields a median survival of 12 to 15 months in the first-line setting and only 6 to 8 months in the salvage setting.
- •
The Cancer Genome Atlas project has provided important insights regarding molecular tumor tissue alterations in bladder cancer.
- •
Emerging data provide promise for a potential role for programmed death 1 and programmed death ligand 1 pathway inhibitors, phosphatidylinositol 3 kinase/mammalian target of rapamycin pathway inhibitors, fibroblast growth factor receptor 3 inhibitors, antiangiogenic agents, epigenetic modulation, and stem cell drivers in selected patients.
- •
Novel clinical trial designs guided by predictive biomarkers based on preclinical data may accelerate therapeutic advances.
Introduction
Despite the high response rates seen in the first-line metastatic setting with cisplatin-based chemotherapy regimens for metastatic urothelial carcinoma (UC), the duration of response is brief and salvage systemic therapy with available agents (vinflunine, taxanes) is marginally active. The limited efficacy of currently used first-line regimens and meager results with salvage chemotherapy translate to poor median survivals of only 12 to 15 months and 6 to 8 months, respectively. In addition, a large proportion of patients, especially elderly patients, do not receive chemotherapy or are ineligible for cisplatin-based combination chemotherapy because of poor performance status, renal dysfunction, and comorbidities, and show poor outcomes with carboplatin-based therapy. The settings of perioperative systemic therapy, combined modality therapy with concurrent radiation, and bacillus Calmette-Guérin (BCG)–resistant non–muscle-invasive bladder cancer (NMIBC) also have suboptimal outcomes. Although no major advances have occurred for the systemic therapy for UC in more than 2 decades, better understanding of tumor biology has identified multiple potential therapeutic targets. This article discusses future directions and highlights emerging promising systemic agents for the treatment of UC.
Introduction
Despite the high response rates seen in the first-line metastatic setting with cisplatin-based chemotherapy regimens for metastatic urothelial carcinoma (UC), the duration of response is brief and salvage systemic therapy with available agents (vinflunine, taxanes) is marginally active. The limited efficacy of currently used first-line regimens and meager results with salvage chemotherapy translate to poor median survivals of only 12 to 15 months and 6 to 8 months, respectively. In addition, a large proportion of patients, especially elderly patients, do not receive chemotherapy or are ineligible for cisplatin-based combination chemotherapy because of poor performance status, renal dysfunction, and comorbidities, and show poor outcomes with carboplatin-based therapy. The settings of perioperative systemic therapy, combined modality therapy with concurrent radiation, and bacillus Calmette-Guérin (BCG)–resistant non–muscle-invasive bladder cancer (NMIBC) also have suboptimal outcomes. Although no major advances have occurred for the systemic therapy for UC in more than 2 decades, better understanding of tumor biology has identified multiple potential therapeutic targets. This article discusses future directions and highlights emerging promising systemic agents for the treatment of UC.
Potential novel therapeutic targets
The Cancer Genome Atlas (TCGA) project recently provided multiple novel insights, although a single dominant tumor driver was not evident ( Fig. 1 ). Chromatin regulatory genes were more frequently mutated in muscle-invasive bladder cancer than in other common malignancies, suggesting a role for targeting epigenetic pathways. Most (76%) tumors harbored an inactivating mutation in 1 or more of the chromatin regulatory genes, and 41% had at least 2 such mutations. Recurrent mutations were observed in genes involved in cell-cycle regulation, chromatin regulation, and kinase signaling. RNA sequencing revealed 4 expression subtypes, including a papillarylike and basal-like/squamouslike subtype. Activating FGFR3-TACC3 fusions and FGFR3 mutations were seen and potential therapeutic targets were noted in 69% of tumors, including targets in the phosphatidylinositol 3 kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR), and mitogen-activated protein kinase (MAPK) pathway. Other studies of bladder cancer have also shown molecular subtypes and have suggested some actionable alterations, including those expressed in tumor initiating or stemlike cells (eg, Stat3, Gli1). Stromal molecules, such as angiogenesis and T-lymphocyte checkpoints, may also represent broad therapeutic targets.
Immunotherapy
Inhibitors of T-lymphocyte checkpoint receptors, especially programmed death (PD-1)/programmed death ligand 1 (PD-L1), are emerging as highly promising agents to induce durable remissions in multiple malignancies. A phase II trial (n = 70) recently showed a high response rate coupled with potential response durability with MPDL3280a, a PD-L1 inhibiting monoclonal antibody (MAb), in the salvage setting. Moreover, higher tumor expression of PD-L1 by immunohistochemistry (IHC) correlated with a robust response rate (2%), although a smaller proportion of patients with no PD-L1 expression also responded (4%). Responses occurred early and the duration of response appeared promising. The toxicity profile in the short term was reassuring with a low rate (3%–4%) of severe toxicities. Aggressive clinical investigation of MPDL3280a is ongoing in a large nonrandomized phase II trial in the salvage setting ( NCT02108652 ), which also accrues to a smaller cohort of cisplatin-ineligible patients as first-line therapy. Furthermore, investigation of pembrolizumab, an MAb targeting PD-1, is planned as a phase III trial in the salvage context following promising results in a phase Ib trial ( Table 1 ). In this phase Ib trial, responses were seen in 24% of 33 pre-treated patients, and many responses were durable.
| Phase | Therapeutic Target | Control Arm | Experimental Arm | Trial Identifier |
|---|---|---|---|---|
| First-line for Cisplatin-eligible Advanced Disease | ||||
| III | VEGF | GC + placebo | GC + bevacizumab | NCT00942331 , CALGB 90601 |
| II | Tubulins | GC | GC + eribulin | NCT01126749 |
| II | Heat shock protein 27 | GC | GC + OGX-427 | NCT01454089 |
| First-line for Cisplatin-ineligible Advanced Disease | ||||
| II | VEGFR, EGFR | GCa | GCa + Vandetanib | NCT01191892 |
| Second or Later Line for Advanced Disease | ||||
| III | PD-1 | Vinflunine or paclitaxel | Pembrolizumab | Not available |
| II→III | Tubulins | Vinflunine | Cabazitaxel | 2012-002826-55 |
| II | VEGFR | Weekly paclitaxel | Pazopanib | 73030316 (PLUTO) |
| II | Tubulins | Paclitaxel | Nab-paclitaxel | NCT02033993 |
| II | Heat shock protein 27 | Docetaxel | Docetaxel + OGX-427 | NCT01780545 |
| II | VEGFR1 or VEGFR2 | Docetaxel | Docetaxel + ramucirumab Docetaxel + icrucumab | NCT01282463 |
| II | FGFR3 | Docetaxel | R3Mab | Not available |
| Switch Maintenance Therapy for Those with Stable Disease after First-line Chemotherapy | ||||
| II→III | Her2, EGFR | Placebo | Lapatinib | NCT00949455 |
| Neoadjuvant Therapy | ||||
| II | VEGFR, FGFR | GC | GC + nintedanib | ISRCTN56349930 |
| Adjuvant Therapy Following Radical Cystectomy | ||||
| II | Her2 | No therapy | DN24-02 (APC-based vaccine) | NCT01353222 |
Ipilimumab, a cytotoxic T-lymphocyte (CTLA) 4–inhibiting monoclonal antibody modulated the immune state in a neoadjuvant trial, and an ongoing phase II trial is assessing the combination of ipilimumab and gemcitabine plus cisplatin (GC) chemotherapy as first-line therapy ( NCT01524991 ). ALT-801, a fusion molecule of interleukin (IL)-2 and a modified T-lymphocyte receptor designed to recognize a p53 epitope in the context of HLA-A*0201, is being evaluated. At first, toxicities were manageable with outpatient administration and induction of an increase of serum interferon gamma, suggesting a favorable immune response. An ongoing phase I trial ( NCT01326871 ) evaluates the combination of ALT-801 (intravenously) plus GC for patients progressing after prior GC for metastatic disease, and another trial ( NCT01625260 ) evaluates ALT-801 plus gemcitabine (both intravenously) for BCG-resistant NMIBC.
Fibroblast growth factor receptor inhibitors
Dovitinib, a multitargeted tyrosine kinase inhibitor (TKI) that inhibits vascular endothelial growth factor (VEGF) and FGFRs, was evaluated as a single agent in a phase II second-line trial. Tumor tissue was evaluated for FGFR3 mutation and patients were stratified by mutational status. Of 44 patients, only 1 objective response was seen in a subject with wild-type FGFR3. However, more potent and selective pan-FGFR inhibition by BGJ398 and JNJ-42756493 has preliminarily shown robust responses in UC harboring FGFR3 mutations, FGFR3/TACC3 translocation, or FGFR2 truncation. Further clinical development of these agents and an Mab targeting FGFR3 are planned in randomized trials that will accrue a selected targeted population (see Table 1 ).
Phosphatidylinositol 3 kinase/AKT/mammalian target of rapamycin pathway inhibitors
Mutations involving the PI3K/mTOR pathway are prevalent in a subset of patients and may represent potential therapeutic targets in UC. Everolimus, an orally administered mTOR inhibitor, has been evaluated following prior platinum-based chemotherapy, alone in a phase II trial, and in combination with pazopanib in phase 1 clinical trial. Despite poor overall activity and outcomes, a small subset of patients whose tumors harbored deletions in the TSC1 (tuberous sclerosis complex 1) gene or activating mTOR mutations experienced exceptionally durable control of disease lasting for more than 1 to 2 years. Everolimus is being evaluated in combination with gemcitabine plus weekly fractionated cisplatin in unselected patients with a baseline creatinine clearance greater than or equal to 40 mL/min in the first-line metastatic setting ( NCT01182168 ). A nonrandomized study of unselected cisplatin-ineligible patients is assessing everolimus alone or with paclitaxel ( NCT01215136 ). In addition, a selective oral PI3K TKI, buparlisib, is being evaluated in unselected patients initially and selected patients with tumors bearing PI3K mutations in the expansion cohort ( NCT01551030 ).
Agents targeting epigenetic pathways
A substantial amount of tumor tissue profiling (notably TCGA data) and preclinical data supports the aggressive evaluation of epigenetic modulators for UC. Inhibition of histone deacetylase (HDAC) results in accumulation of acetylated histones and transcription factors leading to derepression of silenced tumor suppressor genes and antitumor activity. However, romidepsin and vorinostat showed marginal activity. Nevertheless, more potent HDAC inhibitors are undergoing evaluation in patients with tumors harboring epigenetic alterations, such as mocetinostat, an oral second-generation HDAC inhibitor, for tumors bearing inactivating alterations in the histone acetyltransferase genes like EP300 or CREEBB. Azacitidine, a hypomethylating agent, showed significant activity in an autochthonous canine bladder cancer model. Note that epigenetic modulation of tumor tissue may augment the activity of subsequent chemotherapy, as suggested in other malignancies.
Antiangiogenic agents
Bevacizumab
Data support targeting the tumor vasculature in general, and VEGF in particular, in UC. In a phase II trial, bevacizumab, a humanized VEGF Mab, was evaluated in combination with GC as first-line chemotherapy. The overall response rate (ORR) was 72%, including 19% with complete responses (CRs). The median progression-free survival (PFS) of 8.2 months and median overall survival (OS) of 19.1 months suggested a potential improvement compared with GC alone. Thromboembolic events occurred in 21% of patients, although the incidence diminished to 8% after the protocol was amended to administer a lower dose of gemcitabine (1000 mg/m 2 ). Bevacizumab has also been studied in a phase II trial in combination with gemcitabine plus carboplatin (GCa) in patients who were cisplatin-ineligible or deemed to be incurable by cisplatin because of visceral disease. The ORR was 49%, including 6% CR and a median PFS and OS of 6.5 and 13.9 months, respectively. In addition, bevacizumab has been combined with GC and dose-dense MVAC (methotrexate, vinblastine, doxorubicin, cisplatin) preceding radical cystectomy in neoadjuvant nonrandomized trials. However, the nonrandomized design renders them difficult to interpret. The phase III trial (see Table 1 ) comparing GC with either placebo or bevacizumab is expected to complete accrual in 2014, and will provide more definitive data. This important trial also allows maintenance bevacizumab or placebo in patients with at least stable disease following combination chemotherapy.
Sunitinib
Sunitinib, an oral small molecule, multitargeted receptor TKI with activity against VEGF receptor (VEGFR) and platelet-derived growth factor receptor (PDGFR) has been evaluated in multiple settings. As a single agent in the first-line setting in cisplatin-ineligible patients with Eastern Cooperative Oncology Group (ECOG) performance status (PS) 0 to 1 and calculated creatinine clearance less than 60 mL/min, sunitinib yielded an ORR of 8%, median time to progression of 4.8 months, and median OS of only 8.1 months. Low IL-8 levels and tumor contrast enhancement greater than 40 Hounsfield units at baseline were associated with a better clinical outcome, suggesting their potential as biomarkers. When using sunitinib as second-line therapy, the ORR, median PFS, and median OS were 5%, ∼2.3 to 2.4 months, and ∼6 to 7 months. However, a placebo-controlled randomized phase II trial to evaluate a switch maintenance strategy of sunitinib after stability or response to first-line chemotherapy closed prematurely secondary to slow accrual and showed similar suboptimal median PFS (∼2.8 months) and OS (∼10.4 months) in both arms. Substantial toxicities associated with sunitinib were noted in all of these trials. The efficacy of sunitinib in combination with GC was evaluated as preoperative therapy and as first-line therapy for metastatic disease in separate phase II trials, but toxicities forced premature closure of both trials. In a phase II trial, sunitinib seemed to have clinical activity in patients with BCG-resistant NMIBC and increased immune activity.
Pazopanib
Pazopanib, another TKI targeting VEGFR and PDGFR, has been evaluated as a single agent in the heavily pretreated platinum-refractory disease setting in which the ORR was 17%, but with a very low median PFS and OS of 2·6 months and 4·7 months. However, a subset of patients seemed to garner durable benefits: 4 (10%) were progression free after a median follow-up of 19 months. Similar to the first-line sunitinib trial, high baseline IL-8 levels conferred poor outcomes, and early increases during treatment were also associated with poor outcomes. A second phase II single-agent study of pazopanib in the second-line setting was aborted because of futility. Nevertheless, a randomized phase II trial is comparing second-line weekly paclitaxel versus pazopanib (see Table 1 ). A more recent nonrandomized phase II trial combined pazopanib with weekly paclitaxel as second-line therapy and suggested a potential improvement in outcomes with a high ORR and median OS of 10 months. In contrast, another second-line trial combining vinflunine plus pazopanib showed prohibitive toxicities requiring substantial dose reduction of pazopanib to 200 mg once daily. A frontline study of cisplatin-unfit patients treated with gemcitabine plus pazopanib ( NCT01622660 ) is ongoing.
Vandetanib
Vandetanib, a VEGFR and epidermal growth factor receptor (EGFR) TKI, has been studied as salvage therapy after up to 3 prior lines of therapy. In this randomized phase II trial of docetaxel combined with placebo or vandetanib, there was no advantage in outcomes in term of PFS, OS, or responses. A randomized phase II study to evaluate GCa with or without vandetanib for the first-line treatment of cisplatin-ineligible patients is recruiting (see Table 1 ).
Sorafenib
Sorafenib, a TKI targeting VEGFRs and PDGFRs, has shown virtually no activity as a single agent in the first-line and second-line settings. A randomized phase II trial evaluating sorafenib combined with GC was closed prematurely because of slow accrual and showed no improvement in efficacy. Sorafenib in combination with GCa showed excessive hematologic toxicities in a phase II trial and the efficacy was similar to that of GCa. However, sorafenib alone after the completion of GCa plus sorafenib seemed feasible and suggested a durable benefit. A current phase I/II trial is evaluating the combination of sorafenib with vinflunine in the salvage setting ( NCT01844947 ).
Cabozantinib
Cabozantinib is a TKI targeting VEGFR and MET, has shown promising bone metastasis–targeted activity in advanced prostate cancer and is currently under evaluation as salvage therapy for patients with metastatic UC ( NCT01688999 ). Preliminary responses have been observed and associated with low peripheral blood regulatory T cells (T regs ). Moreover, cabozantinib decreased T regs and increased PD-1 expression in T regs , suggesting a potential role for combination or sequencing with immunotherapy.
Other Antiangiogenic Agents
An ongoing 3-arm randomized phase II trial is evaluating docetaxel alone or combined with ramucirumab (VEGFR2 MAb) or icrucumab (VEGFR1 Mab) (see Table 1 ). Aflibercept, a recombinant fusion protein that binds multiple VEGF isoforms and placental growth factor, showed poor activity as a single-agent salvage therapy. TRC105, an antibody targeting CD105, a transforming growth factor beta (TGF-β) coreceptor expressed on the endothelium, is being studied in a phase II trial ( NCT01328574 ). Activin receptor–like kinase (ALK)-1, a TGF-β subclass receptor, is being targeted by PF-03446962 in the second-line setting ( NCT01620970 ). Trebananib is an MAb that targets angiopoietin (Ang)-1 and Ang-2 and has shown promising activity in combination with chemotherapy in a phase I trial. Although the further development of trebananib in UC is not ongoing, a phase II trial of regorafenib, an oral TKI approved in colorectal carcinoma and gastrointestinal tumors with activity against VEGFRs, TIE2, PDGFR, and FGFR, is planned. CEP-11981, a TKI with similar spectrum of antiangiogenic activity including TIE2, had significant preclinical activity. The neoadjuvant setting is also being explored: nintedanib, a TKI targeting VEGFR and FGFR, is being investigated in a randomized phase II trial in combination with GC to evaluate pathologic complete response (pCR) as the primary end point (see Table 1 ). Lenalidomide, an oral immune-modulating and antiangiogenic agent, seems to have antitumor activity in indolent UC preclinical models, and is being evaluated in combination with BCG for BCG-resistant NMIBC ( NCT01373294 ). The combination of lenalidomide and GC was terminated because of toxicities, but another ongoing phase I/II trial is combining lenalidomide with GCa ( NCT01352962 ).
Related posts:
Bladder Cancer
Nonurothelial Bladder Cancer and Rare Variant Histologies
First-line Treatment of Metastatic Disease
Diagnosis and Staging of Bladder Cancer
Perioperative Therapy for Muscle Invasive Bladder Cancer
First-Line Treatment and Prognostic Factors of Metastatic Bladder Cancer for Platinum-Eligible Patients
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
