Emerging therapeutic agents for genitourinary cancers

UC has long been considered an immunogenic tumor [21]. In fact, its immunogenicity has been harnessed as a treatment modality and UC has one of the longest track records of responsiveness to immunotherapy. Bacillus Calmette–Guérin was introduced as a treatment over 40 years ago [22]. Now, immune checkpoint blockade represents the most exciting sphere of emerging therapies for metastatic UC. The objective response rates (ORRs) for the approved post-platinum salvage therapy CPIs (nivolumab, pembrolizumab, avelumab, atezolizumab, and durvalumab) range from 15 to 31% [23]. At the moment, pembrolizumab is the only agent with an OS benefit shown by a randomized phase III study [20, 22].

With regards to CPI as first-line monotherapy in metastatic UC patients, both atezolizumab and pembrolizumab have been examined in patients. Atezolizumab, in the phase II IMvigor210 trial, and pembrolizumab, in the phase II KEYNOTE-052 trial, have both demonstrated clinically meaningful efficacy and objective responses [18, 24]. Both agents are now being studied independently in the phase III setting as monotherapies as well as with combination chemotherapy in previously untreated patients with locally advanced unresectable or metastatic disease. The trials are similarly designed, and the primary endpoints are PFS and OS. Atezolizumab in the IMvigor130 (NCT02807636) and pembrolizumab in the KEYNOTE-361 (NCT02853305) trials are currently underway with highly anticipated results. However, the progress of these trials may be muddled. Preliminary results have shown that the effect of these agents may be less effective than chemotherapy in certain patients, and monotherapy should be limited to patients with high PD-L1 expression. In fact, patients with low PD-L1 levels in the CPI arm of the trials had decreased survival compared to patients who received cisplatin- or carboplatin-based chemotherapy. Patients with low PD-L1 levels are no longer being enrolled in the KEYNOTE-361 or IMvigor130 trials [25, 26].

Novel CPI agents being investigated in the UC model include the IgG2 anti-CTLA-4 monoclonal antibody tremelimumab. Early phase I results in previously treated UC patients receiving combination tremelimumab/durvalumab demonstrated an ORR of 21% with a tolerable adverse event (AE) profile [27]. This combination therapy is now being investigated in the open-label phase III DANUBE trial (NCT02516241). The trial results are highly anticipated and are expected at the end of 2019.

Aside from CPI, alternative immunomodulation strategies in UC have been explored. There are two cytokine-based agonists under investigation at this time in metastatic UC. NKTR-214 is a novel agent being explored in the phase I/II setting. NKTR-214 is an investigational, first-in-class, CD122 preferential agonist that functions as a pegylated recombinant interleukin-2 (IL-2) with cellular effects in activation of CD8⁺ T and natural killer (NK) cells without unwanted expansion of T regulatory (Treg) cells in the tumor microenvironment [28]. The PIVOT-02 study is a multi-cohort phase I trial of NKTR-214 in combination with either nivolumab or ipilimumab/nivolumab therapy. PIVOT-02 includes patients undergoing first-line immunotherapy naïve and platinum-refractory metastatic UC patients (NCT02983045). Preliminary results presented at the 2019 American Society of Clinical Oncology (ASCO) Genitourinary Cancers Symposium were noteworthy for objective responses. The ORR was 48% in efficacy-evaluable patients, with 19% demonstrating a complete response (CR). ORR by immune-related RECIST was 52%. Treatment was well tolerated with only 15% of patients experiencing grade 3 treatment-related adverse events (TRAEs), and no patients experiencing grade 4/5 TRAEs. Of note, the PIVOT-02 trial demonstrates a thought-provoking phenomenon with regards to PD-L1 expression and may meet the urgent unmet need of novel treatments for patients whose tumors lack PD-L1 expression. The impressive ORR and CR were observed regardless of baseline PD-L1 expression. Moreover, 70% of patients who were PD-L1–negative prior to treatment converted to PD-L1–positive expressers after exposure to combination therapy. All PD-L1–positive patients maintained their PD-L1 positivity [29]. These data represent a remarkable breakthrough in immunotherapy treatment as lack of PD-L1 expression remains a barrier to optimal treatment for many patients. It serves as a proof-of-principle in novel approaches to induction of PD-L1 expression. The exact mechanism of PD-L1 modulation remains unclear, but among patients who underwent on-treatment biopsy versus a baseline biopsy, there was influx of CD8⁺ T cells. NKTR-214 may trigger a more vigorous local immune response within the tumor microenvironment [30]. These preliminary trial results have prompted the more expansive phase II PIVOT-10 study evaluating NKTR-214 in combination with nivolumab in locally advanced or metastatic UC cisplatin-ineligible patients with low PD-L1 expression (NCT03785925). An alternative yet similar trial in premise to PIVOT-02 is the phase I PROPEL trial (NCT03138889). PROPEL is investigating atezolizumab in combination with dose escalations of NKTR-214 in patients with platinum-resistant mUC. Another cytokine agonist being explored in the phase II setting is CYT107, a glycosylated recombinant IL-7 agent. Platinum-resistant and cisplatin-ineligible mUC patients are being subject to treatment with intramuscular CYT107 with atezolizumab versus atezolizumab monotherapy (NCT03513952).

Knudson and colleagues have designed a bidirectional fusion protein integrating both CPI and immune cytokine TGFβ inhibition. Such a novel compound serves as the first in a new class of agents that regulate immune suppression in the tumor microenvironment in distinct yet complementary ways [31]. The agent, named M7824, comprises of the extracellular domain of TGFβRII and is linked with the C-terminus of the human anti-PD-L1 heavy chain. Pre-clinical work has been promising, and phase I enrollment is currently open for locally advanced solid tumors (NCT02517398) [32].

A cytokine modulator with early phase efficacy in UC is utomilumab, an investigational fully humanized IgG2 monoclonal antibody which acts as a 4-1BB agonist. 4-1BB is a receptor ubiquitous to T cells (CD4⁺, CD8⁺, NK, and memory T cells) and signals for T cell expansion. In pre-clinical models, utomilumab has demonstrated anti-tumor activity via T cell–mediated immune responses [33]. The phase Ib KEYNOTE-036 trial employing utomilumab with pembrolizumab in advanced tumors including UC revealed no dose-limiting toxicities and an ORR of 26.1% [34]. Further studies investigating utomilumab in UC patients are expected. Another 4-1BB immunotherapeutic under investigation is urelumab, a fully human monoclonal IgG4k antibody agonist of CD137/4-1BB [35]. Downstream effects include tumor necrosis factor (TNF) signaling cascade activation with effect on activated T and NK cells. Urelumab is under investigation in a phase II study in combination with nivolumab as neoadjuvant therapy in cisplatin-ineligible patients (NCT02845323).

Yet another immunomodulating target with a potential role in UC is OX40. OX40 are TNF proteins that are expressed on activated CD4⁺ and CD8⁺ T cells. OX40 signaling promotes T cell proliferation and survival, enhances cytokine production, and modulates cytokine receptor signaling, effectively augmenting the innate and adaptive components of immunity [36]. Additionally, OX40 activation downregulates Treg activity, further amplifying the process [37]. It is no surprise that investigators have developed targeted antibodies to OX40 for cancer therapy. PF-04518600, MOXR0916, and GSK3174998 are all novel agents that are being explored in advanced cancers in combination therapies. NCT02315066 is an early stage dose-escalation trial testing utomilumab with PF-04518600 in patients with advanced cancer, including UC patients. Results revealed no drug-related deaths, dose-limiting toxicities, or suspected unexpected serious TRAEs [38]. Preliminary results show an ORR of only 5.4%, but the stable disease rate was 29.7%. The stable disease rate in the UC cohort was 50% [39]. Although a small sample size, the staggering results have led to the phase I/II trials, JAVELIN Medley and NCT03217747, employing PF-04518600 with either CPIs, immunomodulators, cisplatin, or radiotherapy. Both studies are actively recruiting patients.

The advent of immunotherapy has been a breakthrough in the treatment landscape of UC; however, durable responses are only observed in a minority of patients, and response rates are approximately 20% in the first- and second-line settings and beyond [45], fewer than that benefit from long-term remissions [44]. Similarly, targeted therapies have historically also provided poor response rates and targeted therapy monotherapy has had little success in the metastatic UC model. However, preclinical data suggest that several known anti-angiogenesis agents and tyrosine kinase inhibitors (TKIs) may augment the effects of immunotherapeutics in the tumor microenvironment [46]. As such, there are numerous ongoing trials investigating known and novel agents in conjunction with immunotherapy in UC.

Vascular endothelial growth factor (VEGF) remains an optimal target as studies have shown that elevated urinary and serum VEGF levels in patients with UC suffer poorer prognoses and harbor more aggressive tumors [47, 48]. A VEGF-A inhibitor, bevacizumab, has already demonstrated pre-clinical efficacy in combination with PD-L1 inhibition in RCC [49]. There are two ongoing phase II trials investigating bevacizumab with atezolizumab in cisplatin-ineligible patients and previously untreated mUC, respectively (NCT03133390, NCT03272217). A second agent targeting VEGF is ramucirumab, a monoclonal antibody (mAB) targeting VEGFR₂. Ramucirumab was recently tested in combination with pembrolizumab in a phase I multi-cohort study in UC patients with prior progression on platinum-based systemic therapy. Combination therapy treated patients suffered tolerable TRAEs and demonstrated objective anti-tumor activity [50]. In the phase III setting, ramucirumab has been paired with traditional chemotherapy in 530 patients with post-platinum mUC in the RANGE trial. Remarkably, the patients in the experimental arm with ramucirumab and docetaxel benefited from a mPFS of 4.07 months [95% CI 2.96–4.47] versus 2.76 months [95% CI 2.60–2.96] (hazard ratio [HR] 0.757, 95% CI 0.607–0.943; p = 0.0118). An objective response was achieved in 24.5% (95% CI 18.8–30.3) of patients allocated to ramucirumab and 14.0% (95% CI 9.4–18.6) assigned to placebo. These results are most noteworthy as they represent the first treatment regimen to demonstrate a PFS advantage over chemotherapy in the post-platinum setting. Moreover, these data further validate VEGF₂ inhibition as a therapeutic avenue in metastatic UC [51]. The RANGE trial will likely set a precedent for future trial development. Other strategies towards VEGF inhibition include a novel recombinant EphB4-HSA fusion protein. EphB4-HSA is being investigated in combination with pembrolizumab in previously untreated stage IV UC as a phase II study (NCT02717156). The study is in the recruitment stage. Lastly, cabozantinib is being investigated in a number of trials. Cabozantinib is a small-molecule TKI with target receptors to RET, KIT, AXL, FLT3, MET, and VEGFR₂ that was recently approved for metastatic RCC (mRCC) in the second-line setting after the METEOR trial [52]. Cabozantinib has demonstrated early success in combination with CPI. The UC cohort in a phase I trial of patients with GU malignancies naïve to CPI demonstrated a mPFS of 12.8 months (95% CI 1.8–N/A) and an OS rate of 70.2% (95% CI 44.4–85.8%) [53, 54]. Cabozantinib is also being investigated with other CPIs, including pembrolizumab and atezolizumab, respectively (NCT03534804, NCT03170960).

Nectins represent an interesting and novel therapeutic target for UC. Nectin-4 is a transmembrane polypeptide involved in cell-adhesion and has a role in tumor proliferation and angiogenesis [55]. Translational researchers have used suppression subtractive hybridization on UC pathological specimens and shown high mRNA expression of nectin-4 in bladder cancer [56]. Drug discovery efforts have produced enfortumab vedotin, a novel antibody-drug conjugate (ADC) composed of a mAB to nectin-4 bound to a potent cytotoxic microtubule inhibitor, monomethyl auristatin E. ADCs are a unique class of agents which combine highly specific mABs with toxic drugs [57]. In a phase I dose-escalation study investigating enfortumab vedotin in 68 patients with metastatic UC, the ORR was 41%, and the disease control rate was 72%. These staggering results also included a highly tolerable toxicity profile with only 9% of patients suffering grade 3/4 TRAEs [58, 59]. These data garnered much excitement, and enfortumab vedotin was granted breakthrough therapy status by the FDA for patients with metastatic UC previously treated with a checkpoint inhibitor. Three subsequent phase I–III trials employing enfortumab vedotin were designed soon thereafter, the EV-103 [60], EV-201(NCT03219333), and EV-301 trials [61]. At the 2019 ASCO annual meeting, the results from the single-arm phase II EV-201 trial was reported. Enfortumab vedotin induced a 44% response rate in patients with locally advanced or metastatic UC. Twelve percent of those patients are currently experiencing a complete response. These results are staggeringly similar to the phase I trial results, which strengthens the enthusiasm for the agent. The EV-201 enrolled patients who had been treated with platinum-based chemotherapy and/or checkpoint inhibitors. The mOS was 11.7 months (95% CI 9.1–N/A), mPFS was 5.8 months (95% CI 4.9–7.5), median duration of response (mDOR) was 7.6 months (range, 0.95–11.30+), all with a well-tolerated adverse effect profile [62]. Enfortumab vedotin is now the first novel therapeutic agent to demonstrate clinical benefit in patients who progressed after CPI therapy. The phase III EV-301 and-201 trials are currently underway.

The human epidermal growth factor receptor (HER) family has been widely investigated, and its targeting is embedded as a cornerstone in the treatment of breast and gastrointestinal malignancies [63]. HER2 (Erb2) activation results in tumor cell growth, proliferation, and even chemotherapy resistance [64]. HER2 expression in UC has been well established and UC has one of the highest rates of HER2 expression of any solid tumor [65]. However, the UC patient population has not benefited from HER2 targeting as the data remains unclear to any clinical efficacy in unselected UC patients. A novel ADC has been developed, trastuzumab deruxtecan, and is being investigated in combination with nivolumab in a multicohort phase I trial inclusive of patients with UC (NCT03523572).

The fibroblast growth factor (FGF) pathway is yet another well-elucidated tyrosine kinase (TK) signaling pathway implicated in tumorigenesis and has a high mutational expression rate in UC [16]. As the first TKI approved in UC therapy, the ORR for erdafitinib was 32.3% with 2.3% having a CR in a clinical trial that included 87 patients with advanced bladder cancer with FGFR₂ or FGFR₃ genetic alterations [66]. Major side effects include vision changes associated with retinal disorder and hyperphosphatemia. BGJ-398 is a pan-FGFR inhibitor that has been studied in patients with metastatic UC. In a phase Ib trial inclusive of previously treated metastatic UC patients with FGFR₃ alterations, BGJ-398 demonstrated a disease control rate of 64.2% [67]. Such a dramatic anti-tumor response has prompted development of a multitude of trials investigating anti-FGF therapies in metastatic UC. Most notable is the BISCAY phase I study, an umbrella trial of durvalumab in combination with a potent and selective novel FGFR inhibitor, AZD4547. The trial is inclusive only of patients presenting with FGFR₃ mutations (NCT02546661). Vofatamab (B-701) is yet another novel FGFR₃ inhibitor, and it is being studied in combination with pembrolizumab in the FIERCE-22 international phase I/II trial in metastatic UC (NCT03123055). Rogaratinib (BAY1163877), a novel pan-FGFR inhibitor by Bayer, is being trialed in various solid tumors, including metastatic UC in the FORT-2 trial as combination therapy with atezolizumab (NCT03473756) [68].

Due to the redundancy of anti-angiogenesis targets shared among solid tumors, investigators are able to repurpose agents already developed for other tumors to the RCC model. Brivanib is an investigational, anti-angiogenesis oral TKI previously developed for the treatment of hepatocellular carcinoma, although it currently holds no FDA approval for clinical use in any setting. Brivanib inhibits VEGFR_2, FGFR, and downregulates cyclin D1, Cdk-2, Cdk-4, cyclin B1, and phospho-c-Myc [89]. Brivanib is under investigation in mRCC in a single-arm, phase II study in patients with refractory metastatic disease (NCT01253668). The study is complete and the announcements of results are pending.

Activin receptor-like kinase 1 (ALK) is a TK member of the TGFβ superfamily. Interestingly, its role as a signaling molecule for angiogenesis is independent of VEGF and FGFR signaling [90]. Dual blockade of VEGF and ALK signaling pathways with combination therapies was a novel and intriguing approach to anti-angiogenesis. To this end, Voss and colleagues developed the DART trials, investigating ALK inhibitor dalantercept in combination with axitinib in patients with mRCC after TKI therapy. Phase I results were promising, combination dalantercept and axitinib was well tolerated, the ORR was 25%, and disease control was reported at 57% [91]. However, the recently reported results from the phase II DART trial in which 124 patients were randomized 1:1 to receive axitinib plus dalantercept versus axitinib plus placebo are less encouraging. There has no mPFS benefit in the dalantercept plus axitinib group and the ORR was 19.0% (95% CI 9.9–31.4%) in the dalantercept plus axitinib group and 24.6% (15 of 61 patients; 95% CI 14.5–37.3%) in the placebo plus axitinib group. Although well tolerated, the combination therapy was considered a failure [92]. At this time, Acceleron pharmaceuticals has discontinued development of dalantercept for mRCC [93]. More promising are ongoing trials investigating endoglin inhibition. Endoglin is a homodimeric TGFβ co-receptor that is upregulated in the setting of VHL mutation and HIF1-α overexpression. It is essential for angiogenesis. Investigators have identified the endoglin glycoprotein as a novel non-VEGF angiogenesis pathway that has the potential to complement VEGF-targeted therapy [94]. As such, Choueiri and colleagues have recently employed a chimeric IgG1 mAB targeting endoglin with axitinib in patients with mRCC. The results of a phase Ib trial were recently released, and the combination therapy demonstrated both clinical activity with partial responses in 29% of patients with no dose-limiting toxicity in a VEGF inhibitor-refractory population [95]. A multicenter, randomized phase II trial investigating combination therapy has recently completed accrual (NCT01806064).

Novel emerging therapeutic agents include CCR4, cMET, and HIF2-α inhibitors. CCR4 has molecular implications in angiogenesis, and its inhibition has shown anti-cancer properties [96, 97]. Mogamulizumab is a mAB inhibitor of CCR4, and it is concurrently being investigated in both the phase I and II settings in advanced cancers including mRCC to assess safety and tolerability (NCT02281409) as well as clinical efficacy when in combination with nivolumab (NCT02946671). The identification of HIF accumulation as a sequela of VHL mutation in RCC provides the rationale for drug development towards HIF inhibition [98]. PT2385 is a novel small-molecule inhibitor of HIF2-α. Both phase I and II trials are actively recruiting patients with advanced ccRCC (NCT02293980, NCT03108066). Lastly, we have witnessed cMET inhibition transition to the forefront of mRCC drug development. The objective responses and OS benefit seen with cabozantinib therapy served as a proof-of-principle that cMET may have an in vivo role in mRCC [3]. Four-small molecule inhibitors of cMET are currently under investigation: crizotinib, volitinib, foretinib, and savolitinib. The EORTC 90101 CREATE trial has demonstrated safety and tolerability of crizotinib mRCC patients with MET amplification [99]. Trials including these agents are ongoing (NCT02761057, NCT03091192).

Two highly anticipated phase III trials in the pipeline are KEYNOTE-581/CLEAR and the CheckMate 9ER trial, neither of which have mature data available at this time. KEYNOTE-581/CLEAR is a multicenter, open-label, phase III study evaluating pembrolizumab plus lenvatinib or lenvatinib plus everolimus or sunitinib monotherapy as first-line treatment for mRCC (NCT02811861). Its preceding phase II trial revealed that pembrolizumab plus lenvatinib therapy offered an improved mPFS at 17.7 months (95% CI 9.6–N/A) as well as improved ORR of 66.7% (95% CI 47.2–82.7). The phase III three-armed expansion plans to recruit 735 treatment-naïve patients. The primary endpoint will be PFS with secondary endpoints as ORR, OS, health-related quality of life (HRQoL), and safety profiles. The CheckMate 9ER trial is a two-armed phase III randomized, open-label study exploring nivolumab plus cabozantinib versus sunitinib monotherapy. Interestingly, a recent phase I trial exploring this combination therapy demonstrated impressive anti-tumor activity but enrolled pretreated patients with mRCC [54]. CheckMate 9ER is now exploring the same combination in 630 previously untreated mRCC patients (NCT03141177) [106]. A last anti-PD-1 and VEGF TKI combinatorial regimen brings together pembrolizumab plus cabozantinib. Although these results are from the phase I setting, they are promising and this combination may have an impact on patient care in the future. Previously treated mRCC patients treated with combination therapy demonstrated encouraging early efficacy with an ORR 25% and a clinical benefit rate 87.5%. Enrollment of a phase II dose expansion is now ongoing (NCT03149822) [107].

There remain CPIs approved for other advanced tumors that have yet to establish a role in the treatment of mRCC. Tremelimumab is a CTLA-4 CPI that was examined in combination with sunitinib in a phase I dose-escalation trial with treatment-naïve patients. Of the patients evaluable for response, the ORR was 43% (95% CI 22–66%) and disease stabilization occurred in 33%. However, this study was halted due to unexpected and surprising TRAEs, including acute renal failure andeath [108]. The excitement for tremelimumab in mRCC has since diminished. Notwithstanding, tremelimumab remains under investigation in multiple settings with ongoing phase I trials; neoadjuvant tremelimumab in combination with durvalumab prior to nephrectomy (NCT02762006); and neoadjuvant tremelimumab monotherapy with and without cryoablation prior to nephrectomy (NCT02626130).

The field of immuno-oncology has unveiled a deeper understanding of the immunoreactivity inherent to mRCC, and investigators continue to identify new co-inhibitory ligands implicated in tumor immune evasion [109]. PD-1 and CTLA-4 receptor pathways belong to the B7/CD28 receptor family and have been the foundation for CPI drug discovery. However, our knowledge of the newer co-stimulatory and co-inhibitory pathways within this family remain rudimentary, and more insight into the receptor pathways within this family of receptors will no doubt offer other avenues for augmenting immune responses in the treatment of cancer [110]. Human endogenous retrovirus-H long terminal repeat-associating protein 2 (HHLA2) is a cell membrane and cytoplasmic protein involved in T cell activation and immune checkpoint blockade. Janakiram et al. has labeled HHLA2 as the third group of the B7-CD28 immune checkpoint family after PD-L1 and CTLA-4 [111, 112]. Chen and colleagues have recently demonstrated that HHLA2 has increased expression in ccRCC tumor tissue and that increased expression leads to a remarkably shorter OS and a poorer prognosis [113]. HHLA2 is emerging as a novel target for CPI therapies.