Targeted Therapy in Advanced Urothelial Carcinoma
Targeted Therapy in Advanced Urothelial Carcinoma
ABSTRACT: Urothelial carcinoma (UC) is a common and deadly cancer in the United States. While molecularly targeted therapies have been integrated into the standard-of-care management of other solid tumors in recent years, the use of targeted therapy in UC has lagged behind. Accordingly, the management of advanced disease, along with outcomes, has remained largely unchanged for the past 2 decades. Despite the lack of new agents in the clinic, preclinical and early clinical studies have demonstrated that numerous potentially “targetable” molecular pathways exist, including the epidermal growth factor receptor (EGFR), vascular endothelial growth factor receptor (VEGFR), human epidermal growth factor receptor 2 (HER2/neu), and insulin-like growth factor 1 receptor (IGF1R) pathways. This review focuses on targeted therapies related to these pathways of interest for the treatment of advanced UC, describing the evidence to support further investigation of these approaches. Notably, the identification and validation of new agents will only occur through accrual to urothelial cancer trials designed to answer these questions, which will require the support of the entire urologic community.
Bladder cancer was the fourth most common and eighth most deadly cancer in males in the United States in 2012, with an estimated 55,600 new cases and 17,910 deaths. In females, an estimated 14,580 new cases and 10,510 deaths occurred in the same year. The most common histologic type of bladder cancer, urothelial carcinoma (UC), accounts for more than 90% of cases. At diagnosis, 30% of cases are muscle-invasive, conferring a significant risk of locally advanced or distant metastatic disease. UC is a chemotherapy-sensitive tumor, with chemotherapy response rates around 50% in trials of modern, combination, platinum-based chemotherapy; however, UC often recurs, with a median survival of 14 to 15 months despite chemotherapy in the metastatic setting. Historically, the standard of care in the locally advanced or metastatic setting was MVAC (methotrexate, vinblastine, Adriamycin [doxorubicin], cisplatin) chemotherapy. However, in 2000, a phase III trial by von der Maase et al demonstrated a survival, response rates, and time to progression equal to MVAC using gemcitabine-cisplatin (GC), with an improved toxicity profile compared to MVAC. Since that time, GC has become a new standard-of-care systemic therapy for locally advanced or metastatic UC. Attempts to improve response rates beyond those seen with GC and MVAC have included the addition of a taxane to GC and the administration of MVAC in a dose-dense fashion. Phase II studies have shown activity with a gemcitabine, platinum, and taxane combination that is promising compared to GC alone. A phase III study comparing GC-paclitaxel to GC alone showed better response rates and a 3-month overall survival benefit with GC-paclitaxel, although the latter did not quite reach statistical significance. In addition, a phase III trial of dose-dense MVAC (HD-MVAC) vs conventional MVAC showed a better overall response rate and progression-free survival for HD-MVAC but equivalent median overall survival for the two regimens.
Despite these efforts to identify new uses for and combinations of traditional cytotoxic chemotherapy in UC, survival has improved very little in the last 2 decades. Furthermore, although new-generation biologic/targeted agents have been integrated into the treatment of other cancer types, these therapies have no current use in standard therapy for advanced UC. Notably, UC has a rich array of potentially “druggable” targets, including the vascular endothelial growth factor (VEGF), epidermal growth factor receptors 1 and 2 (EGFR and HER2/neu), and insulin-like growth factor (IGF) pathways, all of which are implicated in the pathogenesis of UC. It must be acknowledged that the failure to integrate biologic agents into the routine care of UC patients is not the result of negative results from large, phase III studies, but of the inability to initiate and complete phase III studies of these agents in UC.
The human epidermal growth factor receptor (HER) family is composed of four receptor tyrosine kinases: HER1 (Erb-B1 or EGFR), HER2 (Erb-B2 or Neu), HER3 (Erb-B3), and HER4 (Erb-B4). These receptors exist in their inactivated state as cell surface monomers; however, when bound by their appropriate ligand, they undergo homo- or heterodimerization, leading to downstream signaling via a number of pathways that ultimately results in cellular proliferation and survival. EGFR and HER2/neu in particular have been implicated in the pathogenesis of invasive UC.
The EGFR receptor is a plasma membrane glycoprotein that dimerizes after ligand activation. This leads to autophosphorylation, resulting in downstream signaling that contributes to cell proliferation, survival, angiogenesis, and metastasis. EGFR is implicated in many cancers, including UC, in which EGFR is overexpressed compared with normal tissue. More than 50% of human UCs overexpress EGFR, and the level of expression correlates with stage, grade, and survival.[6,7] EGFR is naturally expressed in the epithelial layers, and is expressed in the basal cells of the urothelial mucosa, as well as in all layers of the urothelium. The ligand for this receptor, EGF, has been found to induce or stimulate the proliferation of UC cells.[9,10] Dinney et al established an orthotopic model of UC by implanting 253J malignant urothelial cells into the bladder wall of nude mice; the cell line was then altered to overexpress EGFR, which resulted in increased metastasis, supporting the contribution of EGFR to malignant progression. In addition, treatment of 253J cells with an EGFR inhibitor, cetuximab (Erbitux), resulted in a dose-dependent decrease in cell proliferation. Subsequent therapy with cetuximab in an in vivo model has resulted in regression of local tumors and decreased metastasis. In addition to the above-mentioned studies, other in vitro work has supported inhibition of EGFR-expressing cancer cells by additional anti-EGFR monoclonal antibodies.
Various inhibitors of EGFR have been developed, either as monoclonal antibodies (cetuximab, panitumumab [Vectibix]) or as small-molecule receptor inhibitors (erlotinib [Tarceva], gefitinib [Iressa]). Urothelial tumors with a more “epithelial”—as opposed to “mesenchymal”—phenotype have higher EGFR expression and appear to be quite sensitive to EGFR inhibitors. Thus, the use of EGFR blockade with erlotinib is being explored in early-stage, non−muscle-invasive UC as a means of preventing the potential transition to the mesenchymal phenotype that carries a higher likelihood of muscle invasion and metastatic spread (NCT 00749892). However, preclinical work by Bue et al has also shown strong membrane staining for EGFR in metastatic deposits of UC, making EGFR blockade a rational strategy in the metastatic setting as well.
Gefitinib is an orally active selective EGFR tyrosine kinase inhibitor with in vitro activity in UC. The Southwest Oncology Group (SWOG) study 0031 evaluated the role of single-agent gefitinib as second-line treatment for metastatic UC. The study employed a Simon two-stage design and evaluated 6-month progression-free survival. Thirty-one patients were enrolled; there was one partial response, two patients had stable disease (6%), and 81% of the patients had progressed by the first evaluation. The median overall survival was 3 months. The two-stage design allowed for progression to the second step if ≥ 9/30 patients were progression-free at 6 months. Because the majority of patients had progressed by the first radiographic evaluation, the study was stopped after the first step. The most commonly reported toxicities were fatigue and rash. Blockade of EGFR by single-agent gefitinib in this setting is thus not recommended based on these negative results. Of note, no correlation was seen between expression of EGFR and response.
Additional evaluation of gefitinib in the metastatic setting in combination with chemotherapeutic agents was carried out in the Cancer and Leukemia Group B (CALGB) trial 90102. This phase II trial evaluated cisplatin, 70 mg/m2, with gemcitabine, 1000 mg/m2, on days 1 and 8, plus gefitinib, 500 mg daily, for a 21-day cycle in patients with untreated metastatic disease. Out of 54 patients assessed, there were 23 objective responses (overall response rate, 42.6%) and a median overall survival of 15.1 months. The median time to progression was 7.4 months. Although the combination appeared active, the results were not significantly better than historical data for GC alone. Additional clinical studies of gefitinib include an ongoing phase II trial of docetaxel compared to docetaxel plus gefitinib as maintenance therapy for patients already treated with first-line chemotherapy. The primary endpoint of this trial (NCT00479089) is progression-free survival at 9 months from the start of consolidation.
Erlotinib is a reversible oral kinase inhibitor of EGFR approved for the treatment of non−small-cell lung cancer. This drug was also evaluated in the setting of neoadjuvant therapy for UC. This phase II trial evaluated erlotinib dosed at 150 mg daily as neoadjuvant treatment for muscle-invasive UC. Twenty patients were enrolled to receive 4 weeks of preoperative therapy, with subsequent surgical resection and potential maintenance therapy for patients with organ-confined disease. The primary endpoint was the rate of pathologic complete response. At the time of cystectomy, 25% of patients had no residual disease, 35% had down-staging to less than pT2 disease, and 75% had organ-confined disease. Twelve patients had organ-confined disease at surgery and continued on adjuvant erlotinib for a mean of 29 weeks. At 24.8 months of follow-up, 10 patients were disease-free and 5 had received cytotoxic chemotherapy for node-positive disease. The most common adverse event was a rash, seen in 15 of the 20 patients, grade ≥ 3 in 4 of those patients. The treatment was otherwise well tolerated and appeared to have some activity, but this neoadjuvant approach has not been explored further in phase III trials. Another phase II trial (NCT 00749892) is currently evaluating a similar role for erlotinib in the neoadjuvant setting; this trial may offer additional information on the merits of future exploration of this agent, or of erlotinib in combination with other agents in this setting.
Cetuximab is a human/murine chimeric monoclonal antibody against EGFR that is approved for treatment of colon cancer and head and neck cancer. Preclinical data suggesting activity led to phase I and II clinical trials in patients with UC. A randomized, open-label, noncomparative phase II study evaluated cetuximab with and without paclitaxel in metastatic UC that had progressed after a previous line of chemotherapy. Thirty-nine patients were evaluated, with 11 in the single-agent arm and 28 in the combination arm. The single-agent arm was closed for futility when 9 of 11 patients had progressed by the first disease evaluation at 8 weeks. The combination arm accrued 28 patients who were treated with 4-week cycles of cetuximab, 250 mg/m2, given with paclitaxel, 80 mg/m2, per week. The overall response rate was 25%, with three complete responses and four partial responses. Median progression-free survival was 16.4 weeks, and median overall survival was 42 weeks. Seventeen of the 28 patients had visceral metastasis, with a median progression-free survival of 16.1 weeks. The most common treatment-related adverse events were rash, fatigue, and hypomagnesemia. This study concluded that single-agent cetuximab is inactive in advanced UC, but the responses seen in the combination arm suggest a signal in combination with paclitaxel. The combination arm met the study endpoint of improved progression-free survival compared with historical controls; however, further studies have not been planned at this time. This combination may be interesting in a comparative trial. Results of a randomized phase II trial of GC given with or without cetuximab in patients with unresectable, locally recurrent, or metastatic UC were recently reported. Patients received gemcitabine, 1000 mg/m2, on days 1, 8, and 15, with cisplatin, 70 mg/m2, on day 1. Those in the cetuximab-containing arm also received cetuximab, 500 mg/m2, on days 1 and 15 of a 28-day cycle. Patients were then continued on monotherapy with cetuximab after 4 to 6 cycles. Thrombotic complications prompted a protocol alteration to lower the gemcitabine dose in the combination arm. Of the eligible patients, 28 were randomly assigned to the standard therapy arm and 56 to the cetuximab-containing arm. Progression-free survival and median survival were not improved in the experimental arm. In addition, the combination arm contained more adverse events, including more thromboembolism (6.9% vs 18.6% in the cetuximab arm), hyponatremia (3.5% vs 10.2%), rash (0% vs 28.8%), and fatigue—and two deaths. Thus cetuximab has not clearly improved upon current therapies and does appear to carry significant toxicity.