Metastatic Renal Cell Carcinoma: Recent Advances in the Targeted Therapy Era

EUROPEAN UROLOGY 56 (2009) 959 971 available at www.sciencedirect.com journal homepage: www.europeanurology.com Review Kidney Cancer Metastatic Renal Cell Carcinoma: Recent Advances in the Targeted Therapy Era Giuseppe Di Lorenzo a, *, Riccardo Autorino b, Cora N. Sternberg c a Cattedra di Oncologia Medica, Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica, Università degli Studi Federico II, Napoli, Italy b Clinica Urologica, Seconda Università, Napoli, Italy c Department of Medical Oncology, San Camillo Forlanini Hospital, Rome, Italy Article info Article history: Accepted September 1, 2009 Published online ahead of print on September 8, 2009 Keywords: Metastatic renal cell cancer Targeted therapies Abstract Context: The treatment of metastatic renal cell carcinoma (mrcc) has recently evolved from being predominantly cytokine based to being grounded in the use of targeted agents. Objective: To analyse current evidence on the medical management of mrcc. Evidence acquisition: The PubMed and Medline databases were searched for articles published as of 15 July 2009. Only articles published in English were considered. The search terms were metastatic renal cell cancer, targeted therapy, and immunotherapy. Proceedings from the 2000 2009 conferences of the American Society of Clinical Oncology, the American Urological Association, and the European Association of Urology were also searched for relevant abstracts. Evidence synthesis: Sunitinib has recently emerged as a front-line standard of care in mrcc. Temsirolimus is considered a first-line therapy for patients with poor risk features. Bevacizumab/interferon is likely to be the next U.S. Food and Drug Administration approved first-line treatment. The use of sorafenib has moved toward second-line and later therapy. Everolimus was the first agent to show clinical benefit post tyrosine kinase inhibitor failure in a phase 3 study and is considered the standard of care in this setting. Temsirolimus provided benefit to patients with non clear-cell histology. In preliminary results, a favourable risk benefit ratio has been shown with pazopanib and axitinib as first- and second-line treatment. Until combination therapy is clearly shown to be superior to monotherapy, it should be used in the context of a clinical trial. Deciding which is the best sequence to use in mrcc patients remains up to the best judgement of the treating physician. Cytoreductive nephrectomy in the presence of metastatic disease is often indicated as part of an integrated management strategy. Conclusions: Given considerable advances in understanding the biology of mrcc, several new drugs have recently been developed, offering an increasing number of treatment options. A treatment algorithm based on the best available evidence so far can be therefore postulated, though it continues to evolve as data from ongoing trials become available. # 2009 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. Cattedra di Oncologia Medica, Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica, Università degli Studi Federico II, Naples, Italy. Tel. +39 081 7462053; Fax: +39 081 7463660. E-mail address: giuseppedilorenzoncol@hotmail.com (G. Di Lorenzo). 0302-2838/$ see back matter # 2009 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2009.09.002

960 EUROPEAN UROLOGY 56 (2009) 959 971 1. Introduction 2. Evidence acquisition Metastatic renal cell carcinoma (mrcc) is generally resistant to cytotoxic chemotherapy, with response rates (RRs) not exceeding 10% [1,2]. Interferon (IFN) based immunotherapy results in responses of 10 20%, with a median duration of 3 16 mo [3], and a survival advantage for INF over nonimmunotherapy has been reported in randomised trials [4,5]. High-dose immunotherapy with intravenous interleukin-2 (IL-2) showed durable complete responses (CR) in approximately 6% of patients, although this treatment is associated with significant toxicity [6]. An improved understanding of the molecular biology underlying mrcc has led to the development of several new drugs specifically targeting these pathways, and there is now convincing evidence that they provide a clinical benefit. Thus, after the approval of IL-2 and INF in the 1980s, approvals have been given to sorafenib, sunitinib, and temsirolimus in the United States and Europe. In addition, bevacizumab in combination with INF was approved in Europe, and everolimus has demonstrated activity in patients in whom previous treatment with tyrosine kinase inhibitors (TKIs) has failed; everolimus has already been approved in the United States and Europe [7]. Treatment of mrcc has recently evolved from being predominantly cytokine based to being grounded in the use of drugs targeting vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), and mammalian target of rapamycin (mtor) pathways [8,9] (Fig. 1). The aim of this review is to focus on the results of these trials and ongoing efforts to improve treatment of mrcc. It evaluates agents that have been approved in the first- and second-line setting along with their toxicity, placing perspective on the use of sequential and combination therapy. The PubMed and Medline databases were searched for articles published as of 15 July 2009. Electronic earlyrelease publications were also included. Only articles published in English were considered. The search protocol included a free-text query using the terms metastatic renal cell cancer, targeted therapy, and immunotherapy. Proceedings from the 2000 2009 conferences of the American Society of Clinical Oncology (ASCO), the American Urological Association (AUA), and the European Association of Urology (EAU) were also searched for relevant abstracts. 3. Evidence synthesis 3.1. First-line treatment 3.1.1. Sunitinib (SU011248) Sunitinib inhibits the receptor tyrosine kinase (RTK) VEGF receptor 2 (VEGFR2), PDGF receptor (PDGFR), FMS-like tyrosine kinase 3 (FLT-3), and c-kit [10] (Table 1). A dose of 50 mg given orally once per day for 4 wk is the recommended phase 2 dose based on two phase 1 studies [11,12]. The unprecedented efficacy and manageable toxicity profile of sunitinib in the second-line setting in mrcc led to a phase 3 trial of sunitinib versus INF as first-line therapy [13]. This trial (involving 750 patients) demonstrated a significant advantage in objective RR (ORR; 31% vs 6%) and in progression-free survival (PFS; 11 mo vs 5 mo). Overall survival (OS) data were also reported recently [14]. Sunitinib-treated patients had a median OS of 26.4 mo versus 21.8 mo for IFN-treated patients (Table 2). In this study, several patients who were first randomised to receive INF received sunitinib and other active targeted therapy Fig. 1 Biologic agents and their targets in metastatic renal cell cancer. VHL = von Hippel-Lindau; HIF = hypoxia-inducible factor; mtor = mammalian target of rapamycin; VEGF = vascular endothelial growth factor; PDGF = platelet-derived growth factor; TGF-a = tumour growth factor-alfa; VEGFR = vascular endothelial growth factor receptor; PDGFR = platelet-derived growth factor receptor; EGFR = epidermal growth factor; HGF = hepatocyte growth factor.

EUROPEAN UROLOGY 56 (2009) 959 971 961 Table 1 Selected targeted agents demonstrating activity in renal cell carcinoma Agent Target Trial phase Bevacizumab VEGF 3 Sorafenib (BAY 43-9006) VEGFR2, VEGFR3, PDGFR, FLT-3, c-kit, CRAF, wtbraf, V600E BRAF 3 Sunitinib (SU011248) VEGFR2, PDGFR, FLT-3, c-kit 3 Temsirolimus (CCI-779) mtor 3 Everolimus (RAD001) mtor 3 Axitinib (AG-013736) VEGFR1, VEGFR2, PDGFR, c-kit 2 Pazopanib (GW786034) VEGFR1, VEGFR2, VEGFR3, PDGFR 3 VEGF = vascular endothelial growth factor; VEGFR = vascular endothelial growth factor receptor; PDGFR = platelet-derived growth factor receptor; mtor = mammalian target of rapamycin. Table 2 Selected phase 3 studies of targeted therapy in metastatic renal cell carcinoma Ref N Regimen Clinical setting Efficacy [13,14] 750 IFN vs sunitinib First line Median PFS (mo): 5 IFN vs 11 sunitinib ( p < 0.000001) OS (mo) 21.8 IFN vs 26.4 sunitinib ( p = 0.051) [21] 626 Tem vs IFN vs Tem + IFN First line OS (mo) 10.9 Tem vs 7.3 IFN ( p = 0.008) [24,27] 649 Bev + IFN vs IFN First line Median PFS (mo) 10.2 Bev + IFN vs 5.4 IFN ( p < 0.0001) OS (mo) 23.3 Bev + IFN vs 21.3 IFN ( p = 0.129) [25,28] 732 Bev + IFN vs IFN First line Median PFS (mo) 8.5 Bev + IFN vs 5.2 IFN ( p < 0.0001) OS (mo) 18.3 Bev + IFN vs 17.4 IFN ( p = 0.069) [32] 903 Sorafenib vs placebo Second line (post-cytokine) Median PFS (mo) 5.5 sorafenib vs 2.8 placebo ( p < 0.01) [39] 435 Pazopanib vs placebo First line/second line (post-cytokine) Median PFS (mo) 9.2 pazopanib vs 4.2 placebo [36] 410 Everolimus vs placebo Second line (post-tki) Median PFS (mo) 4.0 Ev vs 1.9 placebo IFN = interferon; PFS = progression-free survival; OS = overall survival; Tem = temsirolimus; Bev = bevacizumab; TKI = tyrosine kinase inhibitor; Ev = everolimus. after progression. The above-mentioned crossover could reduce the difference in OS between the two drugs. On the basis of these data, sunitinib has emerged as a front-line standard of care in mrcc. This drug is notable for its superior ORR (31%) compared with the other agents (for example bevacizumab/ifn, which showed a 25.5% ORR). However, this superiority must be balanced against the agent s toxic effects, which include fatigue, diarrhoea, hypertension, and hypothyroidism [15]. Cardiac toxicity has also been reported with sunitinib [15]. Recently, Di Lorenzo et al conducted a retrospective study showing an 18.9% left ventricular ejection fraction decrease temporally related to the administration of sunitinib. Sunitinib also induced grade 3 hypertension in 9.7% of patients. Multivariate analysis suggested that history of coronary artery disease (CAD) and hypertension were the only significant independent predictors of heart failure [16]. Another important side-effect is hypothyroidism [17]. The reported incidence of sunitinib-induced hypothyroidism is 53 85% and 36 46% in retrospective and prospective studies, respectively. Mechanisms of hypothyroidism induced by sunitinib include drug-induced atrophy of the thyroid through inhibition of its vascularisation, druginduced thyroiditis, reduced synthesis of thyroid hormones, progressive depletion of the thyroid s functional reserve, and inhibition of thyroidal iodine uptake. Levothyroxine is the standard treatment for overt or subclinical hypothyroidism [17]. 3.1.2. Temsirolimus (CCI-779) Temsirolimus, a derivative of sirolimus (rapamycin), inhibits mtor, a non-rtk in the PI3K-Akt pathway controlling the translation of specific messenger RNA (Table 1). This mtor activation has multiple downstream effects, including increasing hypoxia-inducible factor 1a (HIF1A) gene expression [18]. Furthermore, reduced phosphatase and tensin homolog (PTEN) expression has been demonstrated in some renal cell carcinoma (RCC) patients [19], and loss of PTEN function results in Akt phosphorylation, with downstream effects on cell growth and proliferation that may be blocked using rapamycin derivatives [20]. This provides a strong rationale for the use of mtor inhibitors in RCC. After preliminary phase 2 results, a three-arm phase 3 trial compared temsirolimus, IFN, and the combination of the two agents as first-line therapy for poor-risk mrcc patients [21] (Table 2). Inclusion criteria were Karnofsky performance status (PS) <80 and at least three of the following poor-risk features: time from diagnosis to first treatment <1 yr, corrected serum calcium >10 mg/dl, lactic dehydrogenase (LDH) >1.5 times the upper limit of normal, haemoglobin less than the lower limit of normal, and multiple organ sites of metastases. Six hundred twenty-six patients were treated, >80% of whom had a Karnofsky PS of 60 70. Two-thirds of participants had undergone prior nephrectomy, and >80% had predominantly clear-cell histology. Temsirolimus was given at 25 mg/m 2 once per week as a single agent, and IFN was given at 3 million U (with an increase to 18 million U) three times per week as a single agent. In the combination arm, temsirolimus was given at 15 mg/m 2 and IFN at 6 million U on the same schedules. The primary end point of the trial was OS. Temsirolimus was generally better tolerated than IFN or the combination. Fatigue was the most common grade 3 4

962 EUROPEAN UROLOGY 56 (2009) 959 971 toxicity and was seen in 12% of patients on temsirolimus, 27% on IFN, and 30% on combination. The most common laboratory abnormality was anaemia, with grade 3 4 toxicity seen in 21% of patients on temsirolimus, 24% on IFN, and 39% on combination. Sixty-nine percent of patients on temsirolimus had at least one grade 3 4 toxicity, in comparison with 85% on INF and 87% on the combination. RRs were similar in all three arms and ranged between 7% and 11%. Median OS was longer in the temsirolimus singleagent arm in comparison with the other two arms (10.9 mo for temsirolimus, 7.3 mo for INF, and 8.4 mo for the combination; hazard ratio [HR]: 0.73, p = 0.0069 for singleagent temsirolimus; Table 2). Overall, temsirolimus as a single agent significantly improved the OS of poor-risk mrcc patients, and it was better tolerated than IFN. Thus, it should be considered a first-line therapy for patients with poor risk features [21]. 3.1.3. Bevacizumab Bevacizumab is a humanised monoclonal antibody that is directed against VEGF (Table 1) [22]. It was one of the first antiangiogenic agents to demonstrate activity against mrcc in a randomised phase 2 trial [23]. Multicenter trials have now established bevacizumab-based therapy as robust in the front-line setting [24,25]. Escudier and colleagues randomised 649 nephrectomised patients with untreated clear cell mrcc to IFN (9 million U subcutaneously 3 times per week) combined with either bevacizumab (10 mg/kg every 2 wk) or placebo [24]. Median PFS was significantly improved in the bevacizumab plus IFN group compared to the control group (10.2 vs 5.4 mo; p = 0.0001). RRs were also significantly improved in the bevacizumab and IFN groups (31% vs 13%; p < 0.0001). The most common grade 3 toxicity was fatigue: 12% in the bevacizumab and IFN arm versus 8% in the control arm. A second multicenter phase 3 trial, conducted in the United States and Canada through the Cancer and Leukaemia Group B (CALGB), was nearly identical in design with the exception that it lacked a placebo infusion and did not require prior nephrectomy [25]. Median PFS was 8.5 mo in patients receiving bevacizumab plus IFN versus 5.2 mo for patients receiving IFN monotherapy ( p < 0.0001). In addition, the ORR was higher in patients receiving bevacizumab plus IFN (25.5%) than in those receiving IFN alone (13.1%, p < 0.0001). Toxicity was greater in the bevacizumab and IFN group, including significantly more grade 3 hypertension (9% vs 0%), anorexia (17% vs 8%), fatigue (35% vs 28%), and proteinuria (13% vs 0%). In addition, a subset analysis revealed a similar benefit with the combination in patients who had an IFN dose reduction because of toxicity [26]. At the ASCO 2009 meeting, the two studies reported final OS results [27,28]. Escudier et al presented findings from the AVOREN trial (first-line bevacizumab plus IFN vs IFN alone in mrcc). There was no difference in OS between the two arms [27]. Rini et al presented results of CALGB trial 90206 showing no difference in terms of OS between the combination of bevacizumab and IFN versus IFN alone (18.3 vs 17.4 mo, p = 0.069) [28] (Table 2). Both studies demonstrated a PFS and an RR as well as a trend towards improved OS in favour of the combination arm. No OS differences were demonstrated because of the extremely high rate of second-line therapies in both studies (55% in the combination arm and 63% in the IFN arm of the AVOREN trial; in the CALGB study, 54% and 63%, respectively). OS in the AVOREN trial was approximately 34% at 3 yr compared to 31% in the CALGB study and 42% in the sunitinib trial. No prespecified subgroup analyses revealed OS advantages because of subsequent therapies. Considering that today all fit patients, after first-line therapy, usually receive subsequent treatments, the OS endpoint is probably less useful than PFS and toxicity. Based on these results, bevacizumab/ifn has been approved by the European Medicines Agency (EMEA) and is likely to be approved by the U.S. Food and Drug Administration as firstline treatment for mrcc. However, it is important to note that, in contrast to the AVOREN study, in the 2009 update of the sunitinib trial, although the patients received second- and third-line treatments, the advantage of sunitinib versus IFN in terms of OS was equally observed [14]. 3.2. Second-line therapy 3.2.1. Sorafenib (BAY 43-9006) Sorafenib inhibits the RTKs VEGFR2, VEGFR3, FLT-3, c-kit, and PDGFR as well as the nonreceptor serine threonine kinases BRAF and CRAF [29] (Table 1). The BRAF and CRAF kinases are members of the RAF/MEK/ERK signalling cascade, which is involved in survival and proliferation of tumour cells and is a therapeutic cancer target [30]. The recommended phase 2 dose of sorafenib was 400 mg orally twice daily on the basis of four phase 1 studies [31]. A phase 3 trial of sorafenib known as the TARGET trial randomised 905 patients with treatment-refractory mrcc to receive either sorafenib 400 mg orally twice daily or placebo [32]. A PFS advantage in the treatment arm of 5.5 mo versus 2.8 mo was observed ( p < 0.000001; Table 2). The median OS was 17.8 mo for patients in the sorafenib group and 15.2 mo for those in the placebo group ( p = 0.146). The median OS was 19.3 mo for sorafenib versus 15.9 mo for placebo ( p = 0.015). This is a trend towards not obtaining the level of significance specified for the interim analysis. At ASCO 2006, Eisen presented results on the impact of crossover on survival. Sorafenib produced a survival advantage of a few months over placebo, and there was a benefit to crossover patients [33]. The TARGET trial represents the best evidence in favour of the use of sorafenib as second-line therapy [32]. These phase 3 data are in contrast to previous results from a randomised phase 2 study of sorafenib versus IFN in the first-line setting of mrcc. Median PFS was 5.7 mo with sorafenib versus 5.6 mo with IFN [34]. The reasons for the lack of significant effect remain unclear, but they might be related to several factors: (1) weaker inhibition of VEGFR compared with sunitinib, (2) patients enrolled in the phase 2 study had bulky disease and may not represent the typical

Table 3 Common toxicities of targeted therapy in renal cell carcinoma and management strategies Toxicity Drug Incidence * Recommendation Fatigue All 30 50% all grades Dose interruption/reduction; eliminate other treatable causes such as anaemia and hypothyroidism; supportive care 5 11% grade 3 or higher Hypertension Sunitinib 17 24% all grades Early and aggressive BP management. All classes of antihypertension agents can be effective. Optimal drug choices Sorafenib 4 8% grade 3 or higher (no hepatic metabolism): Telmisartan (20 80 mg QD), valsartan (80 320 mg QD), atenolol (50 100 mg QD), Bevacizumab 3 12% grade 3 or higher HCTZ 12.5 100 mg QD). Example of possible protocol: Start with HCTZ 12.5 mg QD and increase until maximum/ tolerable dose. If hypertension remains uncontrolled, add another medication (ie, atenolol, valsartan, or telmisartan). Note: Non-dihydropyridine calcium channel blockers (verapamil, diltiazem) are contraindicated. Diarrhoea Hand foot skin reaction Sunitinib, sorafenib Bevacizumab plus IFN-a 27 53% all grades Avoid foods that aggravate diarrhoea (spicy, fatty foods, caffeine), stool softeners, and fibre supplements. Oral rehydration with fluids containing water, salt, and sugar. Loperamide (Imodium) and diphenoxylate (Lomotil). 1 5% grade 3 or higher Sunitinib 20 30% all grades Corticosteroids with no proven efficacy. Treatment delay and adjustments if grade 3. Sorafenib 5 6% grade 3 or higher Mucositis Sunitinib 10 45% all grades Dietary modifications (bland foods). Mucosal coating agents (Orabase, Gelclair). Topical lidocaine solution, Sorafenib 1 6% grade 3 or higher combinations of topical lidocaine and mucosal-coating agents. With fungal infection (oral candidasis): Temsirolimus Everolimus oral fluconazole (Diflucan) 200 mg BID for 7 14 d. Proteinuria Bevacizumab 18% all grades Hold bevacizumab dose for grade 3 or greater and resume when proteinuria is reduced. 7% grade 3 or higher Dyspnoea Sorafenib 8 28% all grades Chest x-ray if respiratory symptoms develop in individuals taking temsirolimus to rule out interstitial Bevacizumab 1 9% grade 3 or higher pneumonitis. Consider steroid treatment. plus IFN-a Temsirolimus 9% grade 3 or higher Everolimus Neutropenia Sunitinib/ 7 72% all grades CBC every 2 wk or daily if febrile neutropenia. Counselling to prevent infections (eg, wash hands). Treatment interruption. bevacizumab plus IFN-a Temsirolimus 0 12% grade 3 or higher Everolimus Thrombocytopenia Sunitinib/ 6 65% all grades Supportive and/or dose modification/interruption. bevacizumab plus IFN-a Temsirolimus 1 % grade 3 or higher Everolimus Elevations of blood glucose, triglycerides, cholesterol Temsirolimus Hyperglycaemia Dietary modifications. Standard medications depending on abnormality. (26 50% all grades/ Everolimus 11 12% grade 3 or higher) Hypertriglyceridemia (27 71%/1 3%) Hypercholesterolemia (24 76%/1 3%) Hypothyroidism Sunitinib 21 85% all grades Thyroid hormone replacement if clinical symptoms and/or biochemical abnormalities. Sorafenib 5 23% grade 3 or higher Levothyroxine: Usual starting dosage for adults >50 yr without cardiac disease or <50 yr with cardiac disease = 25 50 mg QD; for adults > 50 yr with cardiac disease = 12.5 25 mg QD. Dose adjustments are needed every 6 8 wk based on clinical and laboratory parameters. BP = blood pressure; QD = four times daily; HCTZ = hydrochlorothiazide; IFN-a = interferon-alfa; BID = twice daily; CBC = complete blood count. * Based on data from phase 3 trials. EUROPEAN UROLOGY 56 (2009) 959 971 963

964 EUROPEAN UROLOGY 56 (2009) 959 971 patients enrolled in European and US trials, and (3) poor study design or underpowering. Recently, sorafenib received EMEA approval for patients who are treatment refractory to cytokines or patients not eligible for cytokines as first-line therapy. 3.2.2. Everolimus (RAD001) A different mtor inhibitor, everolimus (RAD001), has shown activity in RCC (Table 1). Forty-one patients were treated with RAD001 in a single-arm phase 2 trial at a dose of 10 mg daily orally, of which 37 patients were evaluable for response. Partial responses (PR) were reported in 12 (32%) patients, with 19 (51%) patients having stable disease for >3 mo. Treatment was well tolerated, the most common grade 3 toxicities being anaemia, thrombocytopaenia, hyperglycaemia, and hypophosphataemia in 5 10% of patients [35]. A randomised phase 3 study of everolimus versus best supportive care in patients with mrcc progressing after treatment with sorafenib and/or sunitinib has been reported [36] (Table 2). Everolimus resulted in a statistically significant improvement in PFS (4.0 vs 1.9 mo; p < 0.0001). The most common adverse events were stomatitis, anaemia, and asthenia. Therefore, everolimus is the first agent to show clinical benefit post-tki failure in a phase 3 study, and it should be considered the standard of care in this setting [36]. Everolimus is already EMEA approved in this setting. 3.3. New agents in advanced development 3.3.1. Pazopanib Pazopanib has broad spectrum of kinase inhibition, including VEGFR1 3, PDGFR-A and -B, and c-kit [37] (Table 1). It has been studied in a randomised phase 2 discontinuation study in patients with mrcc who were treatment naive or who had failed one line of treatment, not including a TKI [38]. Two hundred twenty-five mrcc patients were treated (69% previously untreated patients and 84% patients with good/intermediate risk). PFS was 11.9 mo for pazopanib versus 6.2 mo for placebo ( p = 0.0128). Toxic effects included those typical of this class of VEGFR inhibitors (ie, fatigue, diarrhoea, hypertension). A phase 3 trial versus placebo in patients who either received no prior therapy or who failed on prior therapy with cytokines or bevacizumab has completed accrual [39].Atthe 2009 ASCO meeting, Sternberg et al reported these results. Four hundred thirty-five naïve and cytokine-pretreated mrcc patients were randomised to oral pazopanib or placebo (randomisation 2:1 for pazopanib), with PFS as the primary endpoint. There was a PFS advantage for pazopanib (median PFS: 9.2 mo vs 4.2 months, p < 0.0000001; Table 2). Interim analysis showed a prolonged OS with pazopanib. These data are highly immature, as there have not yet been enough events as prespecified (48% of placebo patients received pazopanib after progression, and another 8% received sunitinib). Median survival was 21.1 mo for pazopanib versus 18.7 mo for placebo (HR: 0.73; p =0.02). The treatment was well tolerated, and these findings suggest a favourable risk benefit ratio with pazopanib [39]. A phase 3 GlaxoSmithKline-sponsored trial of pazopanib versus sunitinib in patients with treatment-naïve mrcc has been planned. 3.3.2. Axitinib (AG-013736) Axitinib is an inhibitor of RTKs VEGFR1, VEGFR2, PDGFR, and c-kit (Table 1), with an acceptable side-effect profile consisting predominantly of hypertension and stomatitis [40]. A phase 2 study of axitinib in 52 patients with mrcc all of whom had failed at least one prior cytokine-based therapy has been reported [41]. An RR of 44% was observed, with two (4%) CRs and 21 (40%) PRs. Median time to progression was 15.7 mo, and the median OS was 29.9 mo. Treatment-related adverse events included nausea, fatigue, diarrhoea, hoarseness, and anorexia or weight loss. Hypertension occurred in 30 patients but was successfully treated in 22 patients. Axitinib has also been studied in 62 patients with mrcc progressing on sorafenib [42]. An RR of 21% was reported, all of which were PRs (13 of 62), but 55% of patients experienced some degree of tumour shrinkage. Fourteen patients had received both prior sorafenib and sunitinib, of which nine patients achieved some degree of tumour shrinkage with one PR. A phase 3 trial of axitinib versus sorafenib (AXIS Trial) in patients who have failed one prior therapy is accruing patients. 3.4. Toxicity Although the targeted agents used in the treatment of RCC are reasonably well tolerated, their toxicity on a long-term basis is unknown. Table 3 shows important toxicities with these drugs and provides suggestions for how to manage them. Thyroid function tests should be included in routine toxicity assessment of TKIs under clinical evaluation [17]. The importance of a cardiac monitoring during sunitinib therapy is fundamental, especially in patients with a prior history of hypertension and/or CAD [16]. Some recommendations can be considered with regard to managing patients with risk factors of developing cardiotoxicity [16]. Of course, these recommendations are preliminary and are deemed prudent in view of the present information. First, for patients with grade 1 2 hypertension, sunitinib can be used, adding antihypertensive medications and increasing daily blood pressure surveillance. For patients developing grade 3 hypertension, it is useful to increase antihypertensive therapy, adding a second drug with a different mechanism of action. If grade 3 hypertension disappears (controlled value), the patient may continue with the full dose of sunitinib and continue using antihypertensive agents. If the hypertension remains uncontrolled, it is necessary to stop sunitinib because of the risk of cardiac heart failure; eventually, reintroduction of sunitinib can be considered at a reduced dose (Fig. 2). It is important for uro-oncologists to evaluate the risk benefit ratio for patients with metastatic disease and to consider alternate regimens (immunotherapy, bevacizumab, mtor inhibitors) when risks are disproportionate.

EUROPEAN UROLOGY 56 (2009) 959 971 965 Fig. 2 Management of cardiotoxicity following sunitinib treatment. HP = hypertension; pt = patient; CHF = cardiac heart failure. 3.5. Selection of patients for therapy Given that a minority of patients with mrcc benefit from immunotherapy (including durable CRs in a small number) but that a larger proportion of patients benefit from treatment with TKIs (although CRs are extremely rare), there is a need for markers in order to select patients suitable for either therapy. Carbonic anhydrase IX expression has been shown to predict outcome with IL-2 based immunotherapy [43], and it is conceivable that a (radiographic, pathologic, or molecular) marker of angiogenesis could predict benefit from TKIs that act via antiangiogenic pathways. A classic prognostic biomarker in oncology is PS, which applies to patients with metastatic renal cancer [44]. Many other clinical prognostic biomarkers have been proposed, including age; clear-cell histology; a short interval from diagnosis to the development of metastatic disease (generally 1 or 2 yr); a greater number of metastatic sites; anaemia; hypocalcaemia; elevated LDH levels; thrombocytosis; and markers of increased inflammation, such as erythrocyte sedimentation rate and C-reactive protein [45 48]. These markers have been combined into risk groups, the most common of which is the one proposed by the Memorial Sloan-Kettering Cancer Centre group for INF-based therapy [49] (Fig. 3). Such markers are the subject of several ongoing clinical trials and will hopefully help to select those patients most likely to achieve the greatest clinical benefit [50]. Choueiri and colleagues evaluated 53 patients with papillary or chromophobe mrcc receiving either sunitinib or sorafenib as their initial TKI treatment in five US and French cancer centres [51]. Three (25%) of 12 patients with chromophobe cancer achieved a PR (two treated with sunitinib, one with sorafenib), with a PFS of 10.6 mo. Two (4.8%) of 41 papillary cancer patients achieved a PR (both treated with sunitinib). PFS for sunitinib-treated patients with papillary RCC was 11.9 mo compared with 5.1 mo for sorafenib-treated patients ( p < 0.001). Further prospective studies of these agents are needed in non clear-cell RCC to clarify their future role. The randomised three-arm phase 3 study comparing temsirolimus, IFN, and the combination of the two agents as first-line therapy for poor-risk patients with mrcc included a significant number of patients with non clear-cell histology (20% of 626) [21]. A subset analysis has been performed to determine the effect of temsirolimus versus IFN 3.6. Treatment of non clear-cell renal cell carcinoma The vast majority of patients in published clinical trials treated with new, targeted agents have clear-cell histology. The activity of these agents in non clear-cell histology has therefore not been well evaluated. Fig. 3 Memorial Sloan-Kettering Cancer Centre criteria. Absence of those criteria indicates good prognosis; presence of one or two criteria indicates intermediate prognosis; presence of three or more criteria indicates poor prognosis (presence of more than three criteria is a predictor of short survival). LDH = lactate dehydrogenase; PS = performance status.

966 EUROPEAN UROLOGY 56 (2009) 959 971 Table 4 Angiogenesis and other pathway inhibitors in clinical development for advanced renal cell cancer Inhibitor VEGFR TKIs Company Type Recentin (AZD 2171) Astra Zeneca Small molecule Vandetanib Astra Zeneca Small molecule Pazopanib GlaxoSmithKline Small molecule Axitinib Pfizer Small molecule DAST (BAY 73-4506) Bayer Small molecule AMG706 Amgen Small molecule Telatinib Bayer Small molecule Brivanib Bristol-Myers Squibb Small molecule Angiocept Bristol-Myers Squibb Small molecule ABT-869 Abbott Laboratories Small molecule OSI-930 OSI Small molecule CHIR-258 Chiron Small molecule XEL 820 Exelis Small molecule XEL 880 Exelis Small molecule Integrin inhibitors Volociximab PDL BioPharma/ Biogen Idec Monoclonal antibody Vitaxin MedImmune Monoclonal antibody Cilengitide Merck Small molecule E7820 Eisal Small molecule VEGF targeting Bevacizumab Genentech Monoclonal antibody Aflibercept Regeneron Protein INGN 241 Introgen Therapeutics Adenovirus c-met inhibitors AMG 102 Amgen Monoclonal antibody XEL 880 Exelis Small molecule PF2341066 Pfizer Small molecule MP 470 SuperGen Small molecule VEGFR = vascular endothelial growth factor receptor; TKI = tyrosine kinase inhibitor; VEGF = vascular endothelial growth factor. on OS and PFS in patients with clear-cell or other histology [52]. For patients with non clear-cell histology, those in the temsirolimus group had a longer OS and PFS than those in the IFN group (OS: 11.6 vs 4.3 mo, respectively; PFS: 7.0 vs 1.8 mo, respectively). Thus, temsirolimus provided benefit to the patients irrespective of histology, warranting further study in this important subset of patients. Extensive evidence that c-met signalling is involved in the progression and spread of several cancers and an enhanced understanding of its role in disease have generated considerable interest in c-met and hepatocyte growth factor as major targets in cancer drug development [53]. This interest has led to the development of a variety of c-met pathway antagonists with potential clinical applications. Several c-met antagonists are now under clinical investigation (Table 4). Preliminary clinical results of several of these agents, including both monoclonal antibodies and small-molecule TKIs, have been encouraging. XL880 is a novel inhibitor of RTKs targeting c-met and VEGFR. In an ongoing phase 1 study, PRs were noted in three out of four patients with papillary RCC [54]. This response has led to the initiation of a phase 2 study of XL880 in patients with papillary RCC [55]. Table 4 shows several promising agents now under investigation in clinical trials. 3.7. Combination therapy One way to potentially improve the activity of targeted therapy would be to combine agents targeting different levels in the von Hippel-Lindau (VHL) hypoxia-inducible gene pathway for example, the combination of an mtor inhibitor with an inhibitor of VEGFR. Bevacizumab has been combined with other targeted agents in phase 1 studies that have recommended dosing schedules for future studies. For example, it has been administered in combination with sorafenib in two phase 1/2 studies [56,57]. The combination of the two drugs appears to increase the sorafenib-associated side-effects, generally limiting sorafenib dosing to 200 mg/d, but despite this, preliminary evidence of efficacy was reported. A concern of combination therapy is that overlapping toxicities will result in dose reductions of the agents, negating the benefit of combined therapy. Recently, Feldman et al have evaluated the maximum tolerated dose (MTD) and safety of sunitinib in combination with bevacizumab in patients with advanced RCC [58]. Three cohorts of 3 6 patients were treated with escalated doses of daily oral sunitinib (ie, 25 mg, 37.5 mg, 50mg)for4wkfollowedbya2-wkbreakandfixeddoses of bevacizumab (10 mg/kg) intravenously once every 2 wk. Dose-limiting toxicities were assessed during the first cycle to determine the MTD, and an expanded cohort was treated to obtain additional safety information. Of 26 participants, 25 received treatment at one of three dose levels. Grade 4 haemorrhage, identified as a dose limiting toxicity, occurred in one patient in each of cohorts 2 and 3. The MTD was determined to be sunitinib 50 mg (4 wk on and 2 wk off) and bevacizumab 10 mg/kg, although chronic therapy at these doses might result in grade 3 4 hypertension and hematologic and vascular toxicities. Overall, 48% of patients discontinued treatment because of adverse events. The authors concluded that the combination of sunitinib and bevacizumab caused a high degree of hypertension and vascular and hematologic toxicities at the highest dose level [58]. Until combination therapy is clearly shown to be superior to monotherapy, it should only be used within the context of aclinicaltrial. 3.8. Sequential therapy The development of several targeted agents means that the clinician is now faced with the dilemma of which agent to give and in which order to provide the optimal benefit. It also means that on progression after one targeted agent, other, similar drugs are now available. Tamaskar and colleagues reported an RR of 20% in patients receiving sorafenib or sunitinib after therapy with a variety of antiangiogenic agents [59]. Rini and colleagues reported an RR of 16% to sunitinib in patients who have become refractory to bevacizumab [60]. Axitinib has demonstrated promising activity in sorafenib-refractory patients, with an RR of 21% and some degree of tumour shrinkage in 55% of patients [42].

EUROPEAN UROLOGY 56 (2009) 959 971 967 Prior retrospective studies suggested activity and tolerability with TKIs in the second line after prior TKIs. However, those reports are subject to several biases [61 63]. Despite the absence of prospective information, in clinical practice, sequential therapy with sorafenib and sunitinib has become de facto a standard treatment. Recently, Di Lorenzo et al published a prospective investigation of sequential sorafenib in 52 sunitinibrefractory mrcc patients [64]. The scientific rationale of their trial was that although sorafenib and sunitinib are multitarget TKIs, they have similar, but not identical, targets and a substantial variety of binding specificities among drugs targeting the same kinases. After one cycle, 40 and 12 patients showed stable and progressive disease, respectively, while five cases of PR occurred after two cycles, for an ORR of 9.6%; 76.9% of patients had stable disease for at least one cycle (8 wk), and 80.8% had progressive disease after two cycles (16 wk). Thirteen patients (25%) had some tumour reduction in target lesions. Median time to progression and median OS were 16 wk (range: 8 40) and 32 wk (range: 16 64), respectively (Table 3). Median survival in responders was 48 wk (range: 44 64). Grade 4 toxicity was rare, and grade 2 3 toxicity was well controlled with dose modification [64]. These data are consistent with the hypothesis that drug resistance emerging after initial use of sunitinib is not completely salvaged by sorafenib, but we cannot know whether a different sequence (sunitinib after sorafenib) is active, as suggested by three previous reports [61 63]. An ongoing phase 3 trial (SWITCH trial) will define the best sequence of those two TKIs. Surely, options for overcoming resistance to antiangiogenic therapy may include novel strategies such as blocking the mtor pathway. This strategy has been tested in the phase 3 clinical trial with everolimus, an mtor inhibitor that has been shown to have a PFS advantage when used in patients who had progressed on prior TKI therapy [36]. This strategy is now being tested in clinical trials evaluating the efficacy of temsirolimus versus sorafenib as second-line therapy in patients who have failed first-line sunitinib (NCT00474786) [65] or temsirolimus in combination with bevacizumab following TKI failure (NCT00782275) [66]. Considering the multiple options for treatment of mrcc, it is increasingly important to identify the correct sequence that will improve survival. Further clinical trials, especially comparing a TKI with an mtor inhibitor, will define the best second-line treatment for patients failing first-line sunitinib. Data are not available for either bevacizumab or temsirolimus after TKIs nor for the administration of mtor inhibitors after another mtor inhibitor. Thus, deciding which is the best sequence to use remains speculative. 3.9. Surgical approaches for advanced disease Also in the biologic therapy era, RCC is predominantly a disease for which radical surgery is not only justified but strongly indicated for advanced tumours. Every effort should be made to achieve total surgical excision, presuming that the patient is an appropriate candidate and vital functions are not compromised. Radical nephrectomy in the presence of metastatic disease (called debulking or cytoreductive nephrectomy) is often indicated as part of an integrated management strategy [67]. This approach is supported by the findings of two phase 3 studies in which patients with mrcc were randomly assigned to nephrectomy or no surgery followed by INF-a). In a combined analysis of these trials, median OS was 13.6 mo for nephrectomy plus INF versus 7.8 mo for INF alone (HR: 0.69; p = 0.002) [68,69]. The cytoreductive paradigm is still the predominant one, even though it has not been prospectively tested in patients who receive subsequent targeted therapy [70]. Although cytoreductive nephrectomy seems to benefit many patients with mrcc, it is not curative and should not be performed indiscriminately. Patients most likely to benefit from cytoreduction include those with a substantial proportion of the total tumour burden (eg, >75%) in the involved kidney, good PS with adequate cardiac and pulmonary function, and no symptomatic central nervous system or diffuse liver metastases [70]. In the presence of bone metastases, surgery and biologic agents can be integrated with bisphosphonates, such as zoledronic acid, after evaluation of creatinine clearance. The other main consideration pertains to surgical resectability particularly, the potential for morbidity if the tumour is close to vital structures, encasement of the renal hilum, or other complicating factors [70]. Optimum protocols for integration of surgery and systemic treatments for patients with advanced disease are undergoing investigation. Patients with limited metastatic disease can be considered for metastasectomy, although they represent a minority (2 3%) of cases. Favourable prognostic factors include complete resectability (eg, solitary lung metastasis) and a long interval between initial diagnosis and development of metastases, indicating an indolent course and reinforcing the likelihood that the metastasis is truly solitary. Patients with favourable features can achieve a 5-yr survival of 30% with metastasectomy, and thus surgical resection of metastases should be considered in highly select patients with RCC [71,72]. 3.10. Adjuvant therapy The administration of adjuvant systemic therapy may reduce the risk of distant relapse and hence improve survival. However, no benefit has been demonstrated for adjuvant immunotherapy in RCC [73,74]; in fact, outcomes may be worse in comparison with placebo [75]. No data are reported for the use of kinase inhibitors as adjuvant therapy, and clinical trials that will address this important issue are currently recruiting patients. A large, ongoing, intergroup trial (ECOG-E2805) is randomising high-risk patients who have undergone nephrectomy to receive placebo, sunitinib, or sorafenib for 1 yr. Two other international adjuvant studies are also recruiting. SORCE is a trial of placebo versus 1 yr versus 3 yr of sorafenib in patients with resected RCC at high or intermediate risk of relapse. The S-TRAC trial (sunitinib vs placebo for the treatment of patients at high

968 EUROPEAN UROLOGY 56 (2009) 959 971 risk of recurrent RCC) is randomising high-risk patients who have undergone nephrectomy to receive placebo or sunitinib for 1 yr. 3.11. Neoadjuvant therapy Neoadjuvant therapy with tumour biopsies both before and on treatment may allow insight into the mechanism of action and resistance to kinase inhibitors in vivo as well as possible tumour downstaging. Robert et al have reported the first case of complete histologic remission after neoadjuvant sunitinib treatment. The patient had locally advanced RCC with important extension to the vena cava, and surgical treatment was difficult. Neoadjuvant therapy reduced surgical risks to make treatment acceptable to the patient [76]. Karakiewicz et al described one case of T3b sunitinib neoadjuvant therapy. In this patient, renal tumour decreased from 11 cm to 8 cm, and thrombus downstaged from the atria to the renal vein [77]. Recently Thomas et al, identified patients with RCC treated with sunitinib, sorafenib, or bevacizumab plus IL-2 before tumour resection. Nineteen patients were treated with targeted therapy and subsequently underwent resection. Surgical extirpation involved an open and a laparoscopic approach in 18 and 3 cases, respectively, for locally advanced (8 cases), locally recurrent (6 cases), and metastatic (3 cases) disease. Two patients with extensive bilateral RCC were also treated to downsize the tumours to enable partial nephrectomy. Perioperative complications were noted in 16% of patients. One patient had a significant intraoperative haemorrhage and disseminated intravascular coagulopathy from a concomitant liver resection. An anastomotic bowel leak and abscess were noted postoperatively in another patient who underwent en bloc resection of a retroperitoneal recurrence and adjacent colon [78]. From these observations, we assume that antiangiogenic treatments could change strategies of care for locally advanced or metastatic RCC. It could allow surgical treatment in some contraindicated patients or reduce surgical risks [79]. However, the integration of targeted therapy and surgery requires careful consideration because of the potential for increased perioperative morbidity. The role of antiangiogenic agents as neoadjuvant therapy should be further studied in randomised trials. 3.12. Limitations of Response Evaluation Criteria in Solid Tumours for targeted therapy In medical oncology, the ORR is evaluated using Response Evaluation Criteria in Solid Tumours (RECIST) criteria [80]; however those criteria show limitations when we use a biologic agent and may not be an appropriate end point in the investigation of new cytostatic agents. Size changes for both response and progression remain arbitrary. The measurement of lesions is laborious, and numerous errors occur when obtaining tumour measurements. These errors arise from observer variations of the estimated position of the boundary of lesions. The edges of irregular or infiltrating lesions are often difficult to define, and indeed, some tumours are impossible to measure. The difficulty of distinguishing peritumoural fibrosis from tumour spread further confounds attempts at measurement. RECIST now excludes cystic or necrotic lesions when evaluating response. Although everolimus, sorafenib, and temsirolimus showed some ORR according RECIST criteria, their efficacy was showed by a high degree of tumour burden reduction and especially by an improvement of PFS and OS compared to the standard therapy arm of the clinical trials. In addition, the efficacy of targeted agents is better evaluated with other endpoints, such as PFS and OS, than the RR alone. It is an Fig. 4 Treatment algorithm in metastatic renal cell cancer. IFN = interferon; IL-2 = interleukin-2; VEGFR = vascular endothelial growth factor receptor; TKI = tyrosine kinase inhibitor.

EUROPEAN UROLOGY 56 (2009) 959 971 969 Fig. 5 Most important ongoing phase 3 trials in renal cell cancer. PFS = progression-free survival; OS = overall survival; QoL = quality of life; RR = response rate; DFS = disease-free survival. important point when we design a clinical trial that we do not incorrectly consider a drug ineffective. 4. Conclusions Over the past few years, there have been considerable advances in the understanding of RCC biology. These advances have been translated into the development of several drugs with improved efficacy, of which TKIs have demonstrated the most significant activity. Although not a single patient with mrcc will be cured by targeted therapy, the improvement in OS represents a big advance for a disease with few options until 3 yr ago. Uro-oncologists have a plethora of treatment options for mrcc patients. Sequencing of monotherapies has become the empiric standard. Of course, whether a specific sequence is preferred will require additional evaluation. In addition, combinations of those agents to maximise VEGF pathway inhibition, overcome angiogenic escape mechanisms, and potentially augment antitumor effects are of interest when taken in the context of safety, efficacy, and eventual costs. Overall, the use of these new drugs should be limited to those subsets of patients in whom they have been studied. We report a treatment algorithm based on the best available evidence according to 2009 National Comprehensive Cancer Network guidelines and 2009 EAU guidelines [81,82] (Fig. 4). Our knowledge is limited by the lack of data in certain settings, but it will continue to evolve as new data become available and important ongoing phase 3 trials will be completed (Fig. 5). Author contributions: Giuseppe Di Lorenzo had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Di Lorenzo, Sternberg. Acquisition of data: Di Lorenzo, Autorino. Analysis and interpretation of data: Di Lorenzo, Sternberg. Drafting of the manuscript: Di Lorenzo, Autorino. Critical revision of the manuscript for important intellectual content: Sternberg. Statistical analysis: Di Lorenzo. Obtaining funding: None. Administrative, technical, or material support: None. Supervision: Di Lorenzo. Other (specify): None. Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Dr Sternberg has a consultancy agreement with Pfizer, Sanofi-Aventis, GlaxoSmithKline, and Novartis. Funding/Support and role of the sponsor: None. References [1] Yagoda A, Bander NH. Failure of cytotoxic chemotherapy, 1983 1988, and the emerging role of monoclonal antibodies for renal cancer. Urol Int 1989;44:338 45. [2] Harris DT. Hormonal therapy and chemotherapy of renal-cell carcinoma. Semin Oncol 1983;10:422 30. [3] Horoszewicz JS, Murphy GP. An assessment of the current use of human interferons in therapy of urological cancers. J Urol 1989;142: 1173 80. [4] Medical Research Council Renal Cancer Collaborators. Interferonalpha and survival in metastatic renal carcinoma: early results of a randomised controlled trial. Lancet 1999;353:14 7. [5] Pyrhonen S, Salminen E, Ruutu M, et al. Prospective randomised trial of interferon alfa-2a plus vinblastine versus vinblastine alone in patients with advanced renal cell cancer. J Clin Oncol 1999;17: 2859 67. [6] Rosenberg SA, Yang JC, White DE, Steinberg SM. Durability of complete responses in patients with metastatic cancer treated with high-dose interleukin-2: identification of the antigens mediating response. Ann Surg 1998;228:307 19. [7] Rini BI, Campbell SC, Escudier B. Renal cell carcinoma. Lancet 2009;373:1119 32. [8] Ravaud A, Wallerand H, Culine S, et al. Update on the medical treatment of metastatic renal cell carcinoma. Eur Urol 2008;54: 315 25. [9] Ljungberg B, Hanbury DC, Kuczyk MA, et al. Renal cell carcinoma guideline. Eur Urol 2007;51:1502 10. [10] Mendel DB, Laird AD, Xin X, et al. In vivo antitumor activity of SU11248, a novel tyrosine kinase inhibitor targeting vascular endothelial growth factor and platelet-derived growth factor receptors: determination of a pharmacokinetic/pharmacodynamic relationship. Clin Cancer Res 2003;9:327 37.