Targeted therapy for metastatic renal cell carcinoma: current treatment and future directions

Abstract

An understanding of vascular endothelial growth factor (VEGF) and mammalian target of rapamycin (mTOR) pathways has greatly changed the way metastatic renal cell carcinoma (RCC) is treated. Based on available phase III randomized trials, anti-VEGF agents such as sunitinib, sorafenib, bevacizumab-based therapy, and mTOR-targeted agents such as temsirolimus and everolimus have been used in the treatment armamentarium for this disease. Now that agents directed against these pathways have largely replaced immunotherapy as the standard of care, new questions have emerged and are the subject of ongoing clinical trials. The development of new targeted therapies including axitinib, pazopanib, cediranib, volociximab, tivozanib (AV-951), BAY 73-4506, and c-met inhibitors such as GSK1363089 and ARQ197 may potentially expand the list of treatment options. Sequential and combination targeted therapies are currently under investigation in advanced disease as are adjuvant and neo-adjuvant approaches around nephrectomy.

Introduction

Metastatic renal cell cancer (RCC) is estimated to have caused 13,010 deaths in the United States in 2008 (Jemal et al. 2008). Early trials evaluating chemotherapy in metastatic RCC have demonstrated only low levels of antitumor activity (Motzer and Russo, 2000; Yagoda et al. 1993). Immunotherapy has long been the standard of care for the treatment of metastatic RCC in an attempt to harness the innate immune response of RCC tumors. Interferon (IFN) alpha has produced response rates of up to 15% with modest to no prolongation of overall survival (OS) versus inactive controls (Coppin et al. 2005; Fyfe et al. 1996; Negrier et al. 2007). High-dose interleukin (IL)-2 is noteworthy for a small but real percentage of durable complete remissions, although it can only be applied to a highly selected minority of patients with RCC (McDermott et al. 2005; Yang et al. 2003b). This is because high-dose IL-2 produces significant side effects including capillary leak syndrome that necessitates intensive blood pressure monitoring and the occasional requirement for vasopressors.

With the greater understanding of the molecular mechanisms underlying RCC progression, new agents targeting relevant biologic pathways have been investigated (Cohen and McGovern, 2005), which has translated into success in the clinic. Agents targeting the vascular endothelial growth factor (VEGF) and mammalian target of rapamycin (mTOR) pathways will be summarized. This review will also detail new drug development, combinations of targeted therapy, and peri-operative neo-adjuvant and adjuvant approaches.

Biologic basis of targeted therapy: von Hippel–Lindau

The recognition of hereditary renal neoplasms has catalyzed the discovery of the molecular basis of clear cell RCC. Clinically, the von Hippel–Landau (VHL) syndrome is an inherited constellation of cysts and tumors in the central nervous system (CNS) and abdominal viscera and includes a predisposition to the development of clear cell RCC (Cohen and McGovern, 2005). Patients with VHL syndrome have an aberrant VHL allele on chromosome 3p25 which predisposes them to disease if the second allele is mutated. This is a prime example of the classical two-hit hypothesis for genes with a tumor-suppressor function. VHL is a 213 amino acid protein that polyubiquinates hypoxia-inducible factor 1 alpha (HIF1alpha) which marks it for destruction by the cellular proteosome. Normally, low oxygen conditions allow HIF1alpha to accumulate and bind to HIF1beta thereby creating a complex that transcriptionally activates genes. In patients with aberrant VHL, HIF1alpha is left to accumulate freely without degradation even under normal oxygen conditions and thus the transcription of genes related to glucose metabolism, apoptosis, angiogenesis and endothelial stabilization are abnormally promoted. This disordered response to hypoxia activates over 100 HIF-responsive genes which include growth factors and their receptors such as VEGF, platelet-derived growth factor (PDGF), and transforming growth factor alpha/beta (TGF) (Rini and Small, 2005).

Nonhereditary sporadic clear cell RCCs also exhibit VHL aberrations (Rini and Small, 2005). A single VHL allele deletion occurs in approximately 78.4–98% of sporadic tumors (Banks et al. 2006; Brauch et al. 2000; Gnarra et al. 1994; Kenck et al. 1996; Kondo et al. 2002; Shuin et al. 1994). For the remaining allele, VHL gene mutations are seen in 34–57% while gene inactivation via hypermethylation of CpG-rich DNA islands occurs in about 5–20.4% of clear cell RCC (Banks et al. 2006; Brauch et al. 2000; Clifford et al. 1998; Foster et al. 1994; Kondo et al. 2002). Thus, it is clear that in both hereditary and sporadic cases of clear cell RCC, VHL abnormalities are a key in pathogenesis.

Another downstream effect of the VEGF receptor (VEGFR) pathway is the activation of PI3 kinase and Akt which in turn promote mTOR kinase (Altomare and Testa, 2005). mTOR is a central component of intracellular pathways that promote tumor growth and proliferation, cellular metabolism and is a mediator of the hypoxic response as an upstream activator of HIF1alpha (Hudson et al. 2002). When mTOR and raptor combine to form an activated complex, they phosphorylate and thus activate the eukaryotic translation initiation factor 4E binding protein-1 (eIF-4BP1) and ribosomal S6 kinase (p70s6k). This leads to the synthesis of cellular proliferation proteins such as cyclin D1, angiogenesis mediators such as VEGF, and hypoxia response regulators such as HIF1alpha.

Targeted therapies in practice

The understanding of the biology behind metastatic RCC has converged with the development of new drugs that target downstream effectors of VHL and HIF including VEGF, PDGF and mTOR (Table 1). Several agents have demonstrated significant efficacy in the treatment of metastatic RCC and have been incorporated into the current treatment algorithm (Table 2).

Table 1.

Selected clinical trials of targeted agents in metastatic renal cell carcinoma.

Agent; mechanism of action	Efficacy
	Population and trial arms	RR (%)	PFS (months)	OS (months)
Sunitinib (Figlin et al. 2008; Motzer et al. 2007); TKI of VEGF and related  receptors (oral)	First-line sunitinib versus  IFN	39 versus 8,  p < 0.000001	11 versus 5,  p < 0.001	26.4 vs 21.8, p = 0.051
Sorafenib (Escudier et al. 2007a); TKI of  VEGF and related receptors (oral)	Immunotherapy refractory,  second-line sorafenib  versus placebo	10 versus 2,  p < 0.001	5.5 versus 2.8,  p < 0.01	17.8 vs 15.2,  p = 0.146
Bevacizumab (Escudier et al. 2007b);  VEGF ligand-binding antibody (IV)	First-line bevacizumab +  IFN versus placebo + IFN	31 versus 13,  p < 0.0001	10.4 versus 5.5,  p < 0.0001	23.3 vs 21.3,  p = 0.1291
Bevacizumab (Rini et al. 2009b);  VEGF ligand-binding antibody (IV)	First-line bevacizumab +  IFN versus placebo + IFN	25 versus 13,  p < 0.0001	8.4 versus 4.9,  p < 0.0001	18.3 vs 17.4,  p = 0.069
Temsirolimus (Hudes et al. 2007);  mTOR inhibitor (IV)	Poor risk first-line  temsirolimus versus IFN	8.6 versus  4.8, NS	3.8 versus 1.9,  p < 0.05	10.9 vs 7.3,  p < 0.008
Everolimus (Kay et al. 2009; Motzer et al. 2008); mTOR inhibitor (oral)	TKI refractory second-line  everolimus versus placebo	5 versus 0, N/A	4.9 versus 1.87,  p < 0.0001	14.78 vs 14.39,  p = 0.117

IFN, interferon; mTOR, mammalian target of rapamycin; OS, overall survival; PFS, progression-free survival; RR, response rate; TKI, tyrosine kinase inhibitor; VEGF, vascular endothelial growth factor.

Table 2.

Current treatment algorithm for patients with metastatic renal cell carcinoma (mRCC).

Setting	Patients	Therapy (level 1 evidence)	Other alternatives
First-line therapy	Good or intermediate risk	Sunitinib  Bevacizumab + IFN	HD IL-2  Sorafenib  Observation  Clinical trial
	Poor risk	Temsirolimus	Sunitinib  Clinical trial
Second-line therapy	Cytokine refractory	Sorafenib	Sunitinib  Clinical trial
	Prior VEGF	Everolimus	Targeted therapy not  previously used  Clinical trial

IFN, interferon; VEGF, vascular endothelial growth factor.

VEGFR tyrosine kinase inhibitors

Receptor tyrosine kinases (RTKs) play an integral role in the signaling cascade of VEGF and PDGF (Christensen, 2007). RTKs have an extracellular domain that binds to their respective ligand and an intracellular domain that holds the tyrosine kinase responsible for downstream signaling. Upon ligand binding, the RTKs dimerize or multimerize to induce a conformational change that allows ATP binding resulting in autophosphorylation and transphosphorylation. These tyrosine domains are then able to phosphorylate and activate various proteins in the downstream signal transduction cascade.

Sunitinib

Sunitinib is an oral multikinase inhibitor that blocks VEGFR-1, 2 and 3, PDGFR-B and related RTKs (Mendel et al. 2003). Initial phase II trials of sunitinib in metastatic RCC, involving a total of 169 patients who had failed prior cytokine-based therapy, demonstrated an objective response rate of 45%, a median duration of response of 11.9 months and a median progression-free survival (PFS) of 8.4 months (Motzer et al. 2006, 2007a; Rosenberg et al. 2007). A pivotal phase III randomized controlled trial was subsequently conducted, comparing first-line sunitinib with IFN, which enrolled 750 patients, most of whom (94%) were favourable or intermediate risk Memorial Sloan Kettering Cancer Center (MSKCC) prognostic criteria (one point each for low hemoglobin, elevated serum corrected calcium, elevated LDH, low Karnofsky performance status (<80), and diagnosis to treatment interval <1 year; zero points: favourable prognosis, 1–2 points: intermediate prognosis, 3+ points: poor prognosis) (Motzer et al. 2002). This trial demonstrated a statistically significant advantage in favour of sunitinib for objective response rate (39 versus 8%; p < 0.000001) and the primary endpoint of PFS (median PFS 11 months versus 5 months, hazard ratio (HR) 0.42, p < 0.001) (Motzer et al. 2007b). The median OS of the sunitinib and IFN groups was 26.4 months and 21.8 months, respectively, which was of borderline statistical significance (p = 0.051); however, the trial allowed patients who progressed on IFN to crossover and receive VEGF-targeted therapy and therefore likely mitigated any survival advantage associated with sunitinib (Figlin et al. 2008). Common toxicities associated with sunitinib have included fatigue, hand-foot syndrome, diarrhea, mucositis, hypertension and hypothyroidism. Cardiotoxicity has been reported and thus monitoring may be required in patients with pre-existing heart disease (Witteles et al. 2008). Because of these results, sunitinib has become a standard of care for the first-line treatment of metastatic RCC.

In a population-based retrospective analysis comparing patients treated in the IFN era (n = 131) versus those treated in the sunitinib era (n = 69), the patients treated with first-line sunitinib had an associated doubling in OS compared to those treated with interferon (17.3 versus 8.7 months, p = 0.004) (Heng et al. 2009). When adjusted for MSKCC prognostic profiles, the HR of death for sunitinib versus IFN was 0.049 (p = 0.001). Even those patients classified as having a poor prognosis by MSKCC criteria had a survival advantage (10.7 versus 4.1 months, p = 0.0329), suggesting that use of sunitinib is beneficial in this population as well.

Sorafenib

Sorafenib was initially investigated for its ability to inhibit b-Raf kinase thereby affecting the mitogen-activated protein kinase (MAPK) signaling pathway responsible for downstream proliferation responses. However, it subsequently became clear that sorafenib also had potent activity against VEGFR2, VEGFR3, PDGFB, Flt-3, and c-kit (Ahmad and Eisen, 2004). The Treatment Approaches in Renal Cancer Global Evaluation Trial (TARGET) was the largest study of previously-treated metastatic clear cell RCC, and enrolled 903 patients randomized to oral sorafenib 400 mg twice daily versus placebo (Escudier et al. 2007a). All patients enrolled had favourable or intermediate risk MSKCC prognostic criteria. There was a clear benefit for sorafenib for the study’s primary endpoint of PFS (median PFS 5.5 months versus 2.8 months for placebo, HR = 0.44, p < 0.01). Objective response rate by RECIST criteria was low (10% by investigator assessment) although over 70% of patients had some degree of tumor burden reduction. The common toxicities experienced with sorafenib are similar to sunitinib except that the hand-foot syndrome may be more pronounced and cardiotoxicity and fatigue appears to occur less frequently. Based on these data, sorafenib has been FDA approved and become a standard of care for second-line treatment of mRCC after immunotherapy failure. Sorafenib has also been evaluated in the first-line setting in a randomized phase II trial comparing sorafenib versus IFN in 189 patients with a primary endpoint of PFS. The median PFS for the patients randomized to receive sorafenib was 5.7 months compared to 5.6 months for those receiving IFN (p = 0.504) (Escudier et al. 2009a). Thus, sorafenib has assumed largely a second-line or later role in the treatment of metastatic RCC.

VEGF ligand-directed therapy

Bevacizumab is a recombinant monoclonal neutralizing antibody that binds and neutralizes circulating VEGF. The activity of this agent in RCC was initially identified by small randomized trials (Bukowski et al. 2007; Yang et al. 2003a). A subsequent phase III clinical trial randomized nephrectomized patients with clear cell mRCC to the combination of IFN (three times per week at a dose of 9 MIU for up to 1 year) plus bevacizumab (10 mg/kg IV every 2 weeks) or IFN with placebo until disease progression (the AVOREN study) (Escudier et al. 2007b). The addition of bevacizumab to IFN significantly increased PFS (10.2 versus 5.4 mo, HR = 0.63, p < 0.0001) and objective tumor response rate (31 versus 13%; p < 0.0001). A CALGB phase III trial of similar design confirmed a PFS benefit (8.4 versus 4.9 months, p < 0.0001) and objective response rate benefit (25 versus 13%, p < 0.0001) to bevacizumab plus IFN (Rini et al. 2009b). In both of these trials, over 90% of patients had favourable or intermediate MSKCC prognostic profiles.

The primary endpoint of OS for both of these trials was recently reported showing no statistically significant difference although there was substantial contamination with subsequent VEGF-targeted therapies after progression of initial IFN (e.g. in 62% of IFN patients in the CALGB study) thereby diluting any OS benefit. The median OS for bevacizumab plus IFN versus IFN alone in the AVOREN study was 23.3 versus 21.3 months (p = 0.13) (Escudier et al. 2009b), respectively and for the CALGB study was 18.3 versus 17.4 months (p = 0.069) (Rini et al. 2009b), respectively. Common toxicities associated with bevacizumab include hypertension and proteinuria with rare but serious toxicity including bowel perforation, arterial ischemic events and bleeding. The contribution of IFN to bevacizumab’s effect is unknown as a bevacizumab monotherapy arm was not included in the phase III trials. Nevertheless, it remains another potential treatment option for RCC and US Food and Drug Administration (FDA) approval has been received.

Mammalian target of rapamycin (mTOR) inhibitors

Temsirolimus

Temsirolimus binds to FKBP-12 to create a complex that directly inhibits mTOR to prevent the formation of the activated mTOR-Raptor complex. Temsirolimus was initially evaluated for patients with mRCC in a randomized phase II study of three different dose levels (Atkins et al. 2004). When patients were retrospectively stratified into MSKCC prognostic risk groups, the poor risk group appeared to have a better than expected OS, leading to further evaluation in this population.

The subsequent phase III trial with temsirolimus had a primary endpoint of OS. Six hundred and twenty-six previously untreated patients with poor prognostic criteria were randomized to temsirolimus 25 mg IV weekly, IFN alpha 18 million units (MU) three times a week or temsirolimus 15 mg IV weekly plus IFN 6 MU three times a week (Hudes et al. 2007). To be considered poor risk, patients were required to have three or more of the following adverse risk features: Karnofsky performance status less than 80%, lactate dehydrogenase over 1.5 times the upper limit of normal, hemoglobin below the lower limit of normal, serum corrected calcium more than 10 mg/dl, time from first diagnosis of RCC to start of therapy of less than a year and three or more metastatic sites. Of patients included in this trial, 19% had nonclear cell or unknown histology. Temsirolimus monotherapy demonstrated an OS advantage compared to IFN alpha (10.9 months versus 7.3 months, log rank p < 0.008). The objective response rates were 8.6% for temsirolimus and 4.8% for IFN, which was not statistically significant. The median PFS for the temsirolimus monotherapy arm and interferon arm was 3.8 months (95% confidence interval [CI]: 1.9–2.2) and 1.9 months (95% CI: 3.6–5.2), respectively. Common side effects include fatigue, hypercholesteremia and hyperglycemia. Temsirolimus has become a first-line option for patients with metastatic RCC of any histologic subtype, appropriately applied to patients with poor prognostic criteria.

Everolimus

Another mTOR inhibitor, everolimus (RAD001) has recently been reported to improve progression-free survival in a phase III trial of patients with mRCC who had progressed on sunitinib, sorafenib or both (Kay et al. 2009; Motzer et al. 2008). These patients were randomized to receive either everolimus 10 mg orally daily or placebo and were stratified by the number of previous tyrosine kinase inhibitors (TKI) and MSKCC ‘previously treated’ risk groups (one point each for anemia, hypercalcemia, and Karnofsky performance status <80; 0 points = favorable, 1 point = intermediate, 2 + points = poor risk group). The primary endpoint was PFS and in the everolimus and placebo groups it was 4.9 months and 1.87 months (p < 0.0001), respectively. The PFS benefit was seen in all three MSKCC risk groups. Common side effects included asthenia, anemia and stomatitis. Up to 14% of patients experienced some form of pneumonitis. OS was 14.79 and 14.39 months (p = 0.117) respectively, however crossover to everolimus was permitted in this study. One hundred and six patients randomized to placebo crossed over to receive everolimus after initial progression. For this group, the median PFS was 5.09 months, which is similar to the PFS of the original everolimus group. This is the first agent tested in a second-line trial after initial TKI failure to demonstrate benefit. US FDA approval has recently been granted.

Drugs in development

Axitinib

Axitinib (AG013736) is a small molecule TKI of VEGFR, PDGFR and c-kit. A phase II trial enrolled 62 treatment-refractory patients with RCC that had progressed on sorafenib (Rini et al. 2007). They were treated with oral axitinib 5 mg twice daily. Of 62 patients, 13 (21%) patients exhibited a partial response and the median PFS was 7.4 months. Another phase II trial with axitinib enrolled cytokine-refractory, nephrectomized patients and demonstrated a response rate of 44.2% and a median time to progression of 15.7 months (Rixe et al. 2007). Currently, a large multicenter phase III trial is enrolling patients that progressed on one prior systemic therapy and randomizing them to axitinib or sorafenib (NCT00678392) with PFS as the primary endpoint.

Pazopanib

Pazopanib (GW786034) is another TKI of VEGFR1-3, PDGFR alpha and beta, and c-kit. A randomized discontinuation study was initiated in patients that were treatment naïve (68%) or who had one line of immunotherapy (25%), bevacizumab (3%) or other nontargeted therapy (2%). The first 60 patients demonstrated good disease control rates and thus led the Data and Safety Monitoring Committee to stop the discontinuation randomization phase and allow all patients to continue the drug. Out of the 225 patients enrolled with metastatic RCC, there was a 27% overall response rate by independent review at 12 weeks (Hutson et al. 2007). A double-blind phase III study of pazopanib 800 mg daily versus placebo in a 2:1 randomization of treatment-naive and cytokine-pretreated patients with metastatic RCC was recently reported (Sternberg et al. 2009). The median PFS in the entire cohort was 9.2 for the pazopanib-treated patients versus 4.2 months in those that received placebo control (p < 0.0000001). In the treatment-naïve subpopulation the median PFS was 11.1 versus 2.8 months (p < 0.0000001) for the pazopanib versus placebo groups, respectively. An interim analysis of OS revealed medians of 21.1 months and 18.7 months (not statistically significant) but it should be noted that 48% of placebo patients crossed over to receive pazopanib after documentation of progressive disease which would dilute the OS effect. This trial was undertaken in the era where first-line treatment options such as sunitinib or bevacizumab plus IFN were already standard of care, so it is not known how pazopanib compares to these. Thus, a first-line pazopanib versus sunitinib randomized controlled trial is currently recruiting (NCT00720941) with PFS as the primary endpoint.

BAY 73-4506

BAY 73-4506 is an orally active, potent multikinase inhibitor targeting both tumor cell proliferation and tumor vasculature through inhibition of receptors of tyrosine kinases (VEGFR, KIT, RET, FGFR, and PDGFR) and serine/threonine kinases (RAF and p38MAPK). Previously untreated patients with predominantly clear cell RCC and measurable disease according to RECIST were enrolled in this multicenter, open-label, phase II study. Eligibility criteria included ECOG performance status 0–1, low or intermediate risk MSKCCC prognostic profiles, and adequate bone marrow and organ function. Treatment consisted of BAY 73-4506 160 mg once daily on a 3 weeks on/1 week off schedule. The primary endpoint was overall response rate. Preliminary efficacy data of the 33 patients evaluable for response show a 27% partial response (PR) and a 42% stable disease (SD) rate (Eisen et al. 2009).

Tivozanib

AV-951 (tivozanib) is a potent inhibitor of VEGFR-1, 2 and 3, c-kit and PDGFR kinases. Patients with locally advanced or metastatic RCC of any histology and no prior VEGF-targeted therapy received AV-951 for 16 weeks, after which further treatment was assigned based on response using a randomized discontinuation trial design. Patients with 25% or less tumor shrinkage continued treatment with AV-951, while patients with more than a 25% change from baseline were randomly assigned to receive AV-951 or placebo for 12 weeks (double blinded). The primary endpoints were objective response rate at 16 weeks, percentage of randomly assigned patients remaining progression free at 12 weeks following randomization, and safety profile. Two hundred and seventy-two patients were enrolled with an overall response rate of 25.4% (independent assessment) and a PFS of 8.9–11.8 months (independent assessment and investigator assessment, respectively) (Bhargava et al. 2009). A phase III clinical trial is currently being developed.

Cediranib

Cediranib (AZD2171) is an oral, potent inhibitor of VEGFR1-3, PDGFR beta and Flt-4. In a phase II trial of first-line treatment in patients with progressive, unresectable, advanced metastatic RCC (Sridhar et al. 2007), preliminary results indicated a partial response rate of 38% (6/16). An additional six patients had stable disease and three patients had progressive disease. Mature results from these trials are awaited.

Volociximab

Volociximab is a chimeric monoclonal antibody against α5β1 integrin. This blocks fibronectin in the extracellular matrix from binding to α5β1 integrin which then induces apoptosis of proliferating endothelial cells. Volociximab was investigated in a multicenter phase II study in patients with metastatic clear cell RCC that enrolled 40 evaluable patients. It was well tolerated at 10 mg/kg IV given every 2 weeks. One subject achieved a partial response while 32 subjects had stable disease (Yazji et al. 2007).

Drugs in development for nonclear cell histologies

Conventional clear cell histology is the most common subtype accounting for more than 80% of all RCCs (Cohen and McGovern, 2005) and has been investigated the most comprehensively in clinical trials. The remaining subtypes including papillary (10–15%), chromophobe (5–10%) and collecting duct carcinoma (<1%) have other molecular mechanisms implicated in their pathogenesis (Cohen and McGovern, 2005).

Sunitinib and sorafenib have been described to have activity in papillary and chromophobe RCC. In a report on 53 patients, 41 with papillary and 12 with chromophobe histologies, the response rate, PFS and OS were 10%, 8.6 months, and 19.6 months, respectively. The partial response rate for patients with chromophobe tumours was 25% and PFS was 10.6 months. The partial response rate and PFS for those with papillary RCC was 5% and 7.6 months, with sunitinib-treated patients appearing to have a longer PFS than those treated with sorafenib (11.9 versus 5.1 months, p < 0.001) (Choueiri et al. 2008).

C-met inhibitors such as GSK1363089 (formerly XL880) and ARQ197 have been developed with the knowledge that genetic alterations in papillary RCC are different from those of clear cell RCC. There is frequently a duplication of the c-MET proto-oncogene on chromosome 7 in patients with heritable disease (Delahunt and Eble, 1997). GSK1363089 and ARQ 197 are inhibitors of the c-met receptor tyrosine kinase which is mutated in most heritable papillary cancers and some sporadic papillary cancers. Dose-finding phase I studies have been completed (Eder et al. 2007; Garcia et al. 2007) and these novel agents are currently being studied in patients with metastatic papillary RCC in phase II trials (NCT00345423). Preliminary results for GSK1363089 (an inhibitor of c-met and VEGFR2) given in an intermittent dosing schedule have been reported in a phase II trial enrolling 35 evaluable patients of which four patients had partial responses and 27 had stable disease as best response (Srinivasan et al. 2009). Enrolment continues on this trial to a daily dosing schedule. Of the patients enrolled in the phase I ARQ 197 trial, five had advanced RCC, of which three had stable disease, one had progressive disease, and one has yet to be reported (Garcia et al. 2007).

Sequence of targeted therapy

Currently, we have the fortunate problem of having several agents demonstrating efficacy in the first- and second-line setting, with a number of other small molecule inhibitors that target VEGFR tyrosine kinase being evaluated in mRCC consistently showing activity. With similar mechanisms of action, clinical responses have been observed, including in patients that have previously received TKI therapy (Heng et al. 2007; Shaheen et al. 2006; Shepard et al. 2008). Part of the challenge in moving forward is the lack of understanding of the biologic underpinnings of resistance to the currently approved agents and uniform clinical definitions of what truly constitutes treatment resistance.

A phase II trial has completed enrolment of patients treated with sorafenib after having progressed on sunitinib or bevacizumab with the rationale that there may be noncrossresistant activity of sorafenib in these patients. Out of the 37 patients, 52% experienced tumor shrinkage (defined as ≥5% decrease in tumor measurements) and 14% experienced a true partial response by RECIST criteria (Shepard et al. 2008). This demonstrates that patients that have progressed on one VEGF inhibitor can still respond to another VEGF inhibitor, but whether this demonstrates true noncrossresistance or residual response to reinstitution of a VEGFR directed therapy is unclear.

The impact of these second-line strategies on PFS or OS is currently unclear and can only be determined in randomized trials, which are ongoing. Although the phase III trial of everolimus versus placebo in the second-line setting has made it a second-line standard of care (Motzer et al. 2008), it remains unknown how this strategy of switching from a VEGF targeted therapy to an mTOR inhibitor compares to switching to another VEGF targeted therapy. A phase III randomized trial of patients who progressed on sunitinib is underway comparing temsirolimus (mTOR inhibitor) to sorafenib (VEGFR inhibitor) (NCT00474786) which will help to address this question. Another ongoing phase III trial based in Germany is evaluating the sequence of sunitinib followed by sorafenib or sorafenib followed by sunitinib at progression. The primary objective is to determine if PFS during second-line therapy of sorafenib followed by sunitinib is at least as effective as sunitinib followed by sorafenib (NCT00732914) with an estimated enrolment of 540 patients.

Combinations of targeted therapy

Studies combining targeted therapies are being performed with the known caveat that combinations are associated with high financial cost and risk of increased toxicity due to additive and overlapping side effect profiles. Rational combinations of active agents continue to be evaluated. Currently, combinations of targeted therapy remain experimental and they should only be employed in the context of a clinical trial.

A phase I trial of bevacizumab and sunitinib in a variety of solid tumors led by the Cleveland Clinic reported one unconfirmed partial response in a patient with papillary RCC out of nine evaluable patients (Cooney et al. 2007). Another phase I trial of this combination given exclusively to patients with metastatic RCC reported four of 13 patients with partial responses (Feldman et al. 2007).

A randomized phase II trial studying the combination of bevacizumab and erlotinib (an inhibitor of the epidermal growth factor receptor [EGFR] pathway) versus bevacizumab and placebo revealed no benefit to the combination in terms of overall response rate or PFS (Bukowski et al. 2007).

More recently, combinations of VEGF inhibitors and mTOR inhibitors have been evaluated. A phase I/II trial of temsirolimus and bevacizumab in advanced RCC in receptor TKI-refractory patients was reported. Preliminary data on 32 patients treated with a recommended dose of temsirolimus 25 mg every week and bevacizumab 10 mg/kg every 2 weeks demonstrated an overall response rate of 17.9%, time to progression of 5.3 months and an OS of 14.5 months (Merchan et al. 2009).

A phase I trial of sunitinib plus everolimus suggested that the recommended dose be 37.5 mg daily of sunitinib plus 20 mg weekly of everolimus (Kroog et al. 2009). A 25% overall response rate was observed and dose-limiting toxicities included thrombocytopenia, mucositis and febrile neutropenia. Another phase I trial of sorafenib plus everolimus found the recommended dose to be sorafenib 400 mg twice daily and everolimus 5 mg daily (Harzstark et al. 2009). This was associated with a 29% response rate and dose-limiting toxicities included grade 3 lipase and grade 4 uric acid elevations.

A large, ongoing randomized phase II trial is studying different combinations of bevacizumab, temsirolimus, and sorafenib compared with bevacizumab alone (The BeST Trial (ECOG 2804, NCT00378703) with a primary endpoint of PFS. A planned 360 patients are being randomized to receive one of four treatment arms: bevacizumab alone, bevaciumab with temsirolimus, bevaciumab with sorafenib, or temsirolimus with sorafenib.

A phase II study has been reported that evaluated the activity of bevacizumab (10 mg/kg IV) and everolimus (10 mg daily orally) in patients with advanced RCC. Fifty-nine patients were enrolled: 30 without prior sunitinib or sorafenib therapy and 29 with prior exposure to the agents. For the patients who completed 8 weeks of therapy, an objective response rate was observed in 21% with a minor response or stable disease in 69%. Grade 3 and 4 adverse events included proteinuria (19% of patients), fatigue (9%) and stomatitis (8%) (Whorf et al. 2008). A phase III study evaluating this combination against bevacizumab plus IFN is currently active (NCT00719264) with a planned sample size of 380 patients and a primary endpoint of PFS.

Adjuvant and neo-adjuvant settings

Targeted agents are also being studied in the adjuvant setting for patients with resected high-risk RCC. The Adjuvant Sorafenib or Sunitinib for Unfavorable Renal Carcinoma (ASSURE) intergroup trial randomizes high-risk nephrectomized patients to 1 year of sorafenib, sunitinib or placebo (estimated enrolment: 1332, primary endpoint: disease-free survival (DFS)) (NCT00326898). Other trials such as the phase III sunitinib versus placebo study for the treatment of patients at high risk of recurrent RCC (S-TRAC: estimated enrolment 236, primary endpoint: DFS) (NCT00375674) and the sorafenib versus placebo trial in patients with resected intermediate or high-risk RCC (SORCE: estimated enrolment 1656, primary endpoint: DFS) (NCT00492258) will further help elucidate the effect of these agents in the adjuvant setting.

In the neo-adjuvant setting, an attempt to downstage tumors prior to surgical resection of localized disease is being investigated. Two cycles of sunitinib preoperatively is being studied in a phase II trial to determine the response rate and tumor downsizing prior to nephrectomy (NCT00480935). Preliminary results of a neo-adjuvant sunitinib study in patients with unresectable renal primaries (with or without distant metastases), revealed that three of 13 patients were successfully resected after neo-adjuvant sunitinib. This suggests that there is potentially a role for sunitinib in downstaging tumors to the point where resection is possible (Rini et al. 2009a).

Conclusion

Targeted therapy has changed the landscape of treatment options for metastatic RCC. The disordered response to hypoxia found in RCC has been harnessed and has made VEGF and mTOR-directed pathways a standard of care. This has led to improved response rates and prolonged survival. New drug development, sequencing and combinations of drugs, and the use of these agents in the adjuvant and neo-adjuvant settings are the subject of ongoing clinical trials.

Conflict of interest statement

Authors have received honoraria and/or have served on an advisory board for Pfizer, Novartis, Wyeth, and/or Bayer.