Recent advances in the treatment of advanced renal cell carcinoma: towards multidisciplinary personalized care


Cora N. Sternberg, Department of Medical Oncology, San Camillo and Forlanini Hospitals, Padiglione Flajani, 1st floor, Circonvallazione Gianicolense 87, 00152 Rome, Italy. e-mail:


What's known on the subject? and What does the study add?

With recent improvements in the prognosis for patients with metastatic renal cell carcinoma (mRCC), focus is now shifting towards maximising clinical benefit from targeted therapies. Factors other than efficacy data are increasingly being considered when selecting a treatment strategy, with a view towards optimising clinical outcomes.

This review examines the development and efficacy of targeted agents for the management of mRCC and discusses the potential factors, including resistance mechanisms, sequential therapy, prognostic and predictive markers of response, and adverse event management, that may contribute to successful individually tallored treatment of patients with this disease.

  • • Targeted agents have substantially improved outcomes for patients with metastatic renal cell carcinoma (mRCC).
  • • Treatment focus is now shifting towards achieving a continuum of care such that long-term benefit and extended survival may be achieved through the optimal use of targeted agents.
  • • To achieve this goal, a number of factors which impact on treatment selection and outcomes need to be considered when treating patients with mRCC, such as the optimal sequence of targeted therapies (and the related issue of resistance mechanisms).
  • • Recent advances are also likely to impact on the future treatment of mRCC. Examples include the identification of predictive biomarkers as well as a consideration of patient risk profiles or the safety profile of the selected targeted agent. In addition, attention is focusing on re-defining the role of surgery for the treatment of RCC in the context of targeted therapies.
  • • This review examines the recent and future advances that offer the potential for personalizing treatment by selecting the most appropriate treatment for each patient with a view towards optimizing clinical outcomes.



metastatic renal cell carcinoma


mammalian target of rapamycin


objective response rate


overall survival


progression free survival


renal cell carcinoma


vascular endothelial growth factor


The impact of kidney cancer is illustrated not only by the associated yearly incidence and mortality rates [1] but also by the fact that approximately 20–30% of patients diagnosed with kidney cancer present with metastatic disease [2] and a similar percentage of patients with initially localized disease experience a relapse and develop metastatic disease [3].

Until recently, standard treatment for metastatic renal cell carcinoma (mRCC) was cytokine therapy with interleukin-2 or interferon-α (IFN-α). This was associated with modest survival benefit in a limited subset of patients and with significant clinical toxicities [4]. The use of targeted therapies has significantly improved outcomes for patients with mRCC. At present, in Europe and/or the USA, six targeted agents are approved for first- and second-line treatment of clear cell mRCC: sunitinib, sorafenib, temsirolimus, everolimus, bevacizumab (in combination with interferon) and pazopanib [5–7]. In addition, several new targeted agents, such as axitinib and tivozanib, are also currently in development for the treatment of mRCC [8,9].

Focus is now shifting towards achieving a continuum of care for patients with mRCC such that long-term benefit and extended survival may be achieved through the optimal use of targeted therapies. Advances in the understanding of underlying factors impacting on clinical outcomes also highlight the potential for personalizing treatment for patients with mRCC such that the most appropriate treatment is selected for the right patient. In addition, a consideration of the latest data regarding sequencing of targeted therapies, the potential for further patient selection based on predictive biomarkers, the role of surgery and other factors, such as patient risk profiles and the safety of targeted therapies, all combine to provide an indication of the future direction of mRCC treatment. In this review, we consider these recent advances and examine their potential implications with regards to optimizing treatment for mRCC.


US and European guidelines [5–7] are in agreement on the treatment algorithm for mRCC based on key clinical data (progression-free survival [PFS], overall survival [OS], objective response rate [ORR]) for first and second-line treatment of patients. These data are summarized in Tables 1 and 2[10–20].

Table 1. Key clinical data for the targeted therapies recommended for the first-line treatment of mRCC by the current guidelines [5–7]]
Therapy N Prior nephrectomy (%)MSKCC group favourable/intermediate/poor (%)Median PFS (months) P valueMedian OS (months) P value
  1. *P values by pre-planned unstratified and stratified log-rank test, respectively. †Patients stratified into the poor risk prognostic category on the basis of three of six risk features (five pre-defined Memorial-Sloan Kettering Cancer Centre risk factors plus multiple sites of organ metastases). Reported as time to progression (TTP). NR, not reported; NA, not applicable.

Sunitinib [13,20]3759138/56/611<0.00126.40.051, 0.049*
 vs IFN-α3758934/59/7521.8
vs IFN-α207670/24/763.17.3
Bevacizumab (plus IFN-α) [11,19]32710027/56/9 (NR: 9)10.2<0.00123.30.1291
 vs IFN-α32210029/56/8 (NR: 7)5.421.3
Bevacizumab (plus IFN-α) [15,18]3698526/64/108.5<0.00118.30.069
 vs IFN-α3638526/64/105.217.4
Pazopanib [16,17] (overall)2908939/55/3 (unknown:3)9.2<0.00122.90.224
 vs placebo (overall)1458839/53/3 (unknown: 4)4.220.5
 treatment-naïve patients155NRNR11.1<0.001NRNA
 vs placebo78NRNR2.8  
Table 2. Key clinical data for the targeted therapies recommended for the second-line treatment of mRCC by the current guidelines [5–7]]
Therapy N Prior therapyPrior nephrectomy (%)MSKCC group favourable/intermediate/poor (%)Median PFS (months) P valueMedian OS (months) P value
  1. *P values by pre-planned unstratified and stratified log-rank test, respectively. Patients stratified into the poor-risk prognostic category on the basis of three of six risk features (five pre-defined Memorial-Sloan Kettering Cancer Centre risk factors plus multiple sites of organ metastases). Reported as time to progression (TTP). NR, not reported; NA, not applicable.

Sorafenib [10]451Cytokine-based or radiotherapy9452/48/05.5<0.00117.80.029
 vs placebo4529350/49/0 (<1 unknown)2.814.3*
Everolimus [14] (overall)277Sunitinib and/or sorafenib; prior immunotherapy, chemotherapy, hormone therapy and radiotherapy also permitted9729/56/144.9<0.00114.80.162
 vs placebo (overall)1399628/57/151.914.4
 refractory to sunitinib124NRNR3.9<0.001NRNA
 vs placebo601.8
 refractory to sorafenib81NRNR5.9<0.001NRNA
 vs placebo432.8
Pazopanib [16] cytokine pre-treated patients135CytokineNRNR7.4<0.001NRNA
  vs placebo67 NRNR4.2


Despite the clinical efficacy observed with the targeted therapies, there are still a number of patients who do not achieve benefit from treatment (present with intrinsic resistance to treatment). Others develop resistance following an initial response to treatment [21]. In this context, consideration of factors such as the optimal sequence of targeted therapies (and the related issue of resistance mechanisms) and the stratification of patients into groups most likely to benefit from selected targeted agents is important as these are all likely to shape the treatment of mRCC in the near future.


Despite complete responses being rarely observed with targeted therapies [22], it is still possible to achieve substantial clinical benefit with targeted agents. In the first-line setting, sunitinib, bevacizumab in conjunction with IFN-α and pazopanib have all demonstrated efficacy for the treatment of patients with good and intermediate prognostic risk [6,7]. Similarly, temsirolimus has demonstrated clinical benefit in treatment-naive patients with poor prognostic risk [6,7]. In view of this, appropriate therapy management strategies are useful for maintaining patients on treatment for as long as possible in order to optimize clinical outcomes. In addition, as there is a small subset of patients who do not appear to benefit from treatment with targeted therapies and another who develop resistance [21], it is important that rational selection of first-line treatment includes a consideration of subsequent lines of therapy, and an understanding of the mechanisms of resistance involved [22].

One approach towards maximizing the potential of targeted therapies and extending clinical benefit over the longer-term is to use sequential therapy. This is now routinely used in clinical practice [22]. Sequencing can include selecting subsequent agents with similar mechanisms of action to that of the agent used for initial treatment or selecting an agent with a different mechanism of action. The rationale supporting each of these options includes a consideration of the biology of RCC and the various signalling pathways involved. As such, there may be advantages attached to selecting a particular sequence over another; that is, one vascular endothelial growth factor (VEGF) inhibitor followed by another, vs a VEGF inhibitor followed by a mammalian target of rapamycin (mTOR) inhibitor. This is further supported by observations that cross-resistance is not inevitable between targeted therapies. Indeed, non-cross resistance has been observed between the tyrosine kinase inhibitors [23].

Data from several studies are becoming available demonstrating the benefit of sequential therapy. In addition, this is under investigation in several ongoing clinical studies. The RECORD-1 phase III trial [14] provided support for sequencing with an mTOR inhibitor, everolimus, after a tyrosine kinase inhibitor, demonstrating benefit in the second- and third-line settings. This led to everolimus being approved for the treatment of patients refractory to prior treatment with VEGF-targeted therapies. Similarly, in a small phase II trial, benefit has also been observed when treating with one tyrosine kinase inhibitor after another, i.e. axitinib demonstrated anti-tumour efficacy in patients with prior sorafenib treatment [24]. Recently available data from the phase III AXIS trial [25] demonstrated significantly higher PFS (median, 6.7 months vs 4.7 months, respectively) and ORR (19% vs 9%, respectively) for axitinib when compared with sorafenib for the second-line treatment of mRCC patients. In this study, the majority of patients had received prior therapy with sunitinib (54%) which selectively inhibit VEGF receptors [25]. Thirty-five percent of the patients included in the study had received prior cytokine treatment. Analysis of PFS according to prior treatment indicated greater benefit from axitinib treatment when compared with sorafenib (prior sunitinib treatment, 4.8 months vs 3.4 months, P= 0.011 and prior cytokine treatment, 12.1 months vs 6.5 months, P < 0.001) [25].

Data from retrospective analyses also indicate that sequencing of one VEGF inhibitor after another provides clinical benefit as does treating with an mTOR inhibitor following prior treatment with a VEGF inhibitor [26,27]. Support for such sequencing strategies is also provided by descriptions of various patient case studies, including one demonstrating disease-free survival of 48 months in a patient treated sequentially with sunitinib, everolimus, sorafenib and temsirolimus [28]. In addition, it is also possible that re-challenge with the same therapeutic agent will extend patient survival. In a retrospective analysis, patients re-challenged with sunitinib following prior treatment with sunitinib and other intervening therapies achieved median PFS of 7.2 months [29].

Several planned and ongoing studies are investigating optimal sequencing of available and new therapeutic agents to clarify further the treatment strategy for patients with mRCC (Table 3) [30–36].

Table 3. Planned and ongoing studies examining optimal sequencing of available and new targeted agents for the treatment of patients with mRCC [30–36]]
Clinical studyNumber of patients (N)Study phaseTreatment armsPrimary endpoint
SWITCH (NCT00732914): study comparing sequential treatment with sunitinib vs sorafenib as first- and second-line therapy346IIISunitinib followed by sorafenibProgression-free survival
Sorafenib followed by sunitinib
PISCES (NCT01064310): study comparing patient preference for pazopanib vs sunitinib in a treatment-naïve patients in a cross-over study161IIIPazopanib followed by sunitinibPatient preference
Sunitinib followed by pazopanib
Tivo-1 (NCT01030783): study comparing tivozanib (AV-951) vs sorafenib as first-line and second-line therapy500IIITivozanibProgression-free survival
AXIS (NCT00920816): study comparing axitinib vs sorafenib as second-line therapy723IIIAxitinibProgression-free survival
INTORSECT (NCT00474786): study comparing temsirolimus vs sorafenib in sunitinib-refractory patients510IIITemsirolimusEfficacy and safety
RECORD-3 (NCT00903175): study comparing sunitinib versus everolimus as first- and second-line therapy390IIEverolimus followed by sunitinibProgression-free survival
Sunitinib followed by everolimus
START (NCT01217931): study comparing six different sequences of targeted agents240IIPazopanib followed by bevacizumabTime to overall treatment failure
Pazopanib followed by everolimus
Everolimus followed by bevacizumab
Everolimus followed by pazopanib
Bevacizumab followed by pazopanib
Bevacizumab followed by everolimus


The role of surgery, and more precisely, the timing of surgery, is also under investigation in the context of the treatment for patients with primary metastatic disease. Cytoreductive therapy was shown to provide overall survival benefit in patients treated with IFN-α[37]. However, this survival benefit has not yet been demonstrated conclusively in the context of targeted therapies. In the case of patients with mRCC, tumour nephrectomy or metastasectomy are recommended by current guidelines for patients suitable for surgery and who have a good performance status and low metastatic burden [6,7].

Nephrectomy continues to be used routinely in the treatment of patients with mRCC. Potential benefits include reduction of tumour burden and alleviation of symptoms, additional information on histology and fewer complications. Some limited data support the use of nephrectomy in the context of targeted therapies. In a retrospective study, complete nephrectomy was associated with better overall survival [38]. However, there are also data suggesting that nephrectomy may not provide additional benefit. In a subset analysis of the phase III temsirolimus study, nephrectomy had no impact on overall survival of patients with poor prognostic risk [12,39].

The ongoing CARMENA study (NCT00930033) will help to define the role of surgery in the context of targeted therapies. This study compares OS in patients with favourable or intermediate prognostic risk receiving sunitinib treatment (no prior surgery) with those who received nephrectomy followed by sunitinib treatment [40]. Another study that may define the role of surgery is focusing on determining whether a particular subset of mRCC patients is likely to benefit from nephrectomy. The SURTIME (EORTC 30073; NCT01099423) study focuses on presurgical treatment with sunitinib which may help to select out patients progressing rapidly at metastatic sites and who may not have an additional benefit by nephrectomy [41]. This study compares PFS in patients who receive a nephrectomy followed by sunitinib 4 weeks after the surgery with patients treated with three courses of sunitinib (schedule 4/2; 4 weeks on treatment followed by 2 weeks off) prior to nephrectomy and followed by resumption of sunitinib treatment again. The study aims to determine if immediate vs deferred nephrectomy impacts on disease control in patients with synchronous, resectable, mRCC [41].

The use of targeted agents around nephrectomy also remains under investigation by several studies in the non-metastatic setting. In these patients, surgery remains the mainstay of therapy and the precise role of targeted agents is less clear. In this instance, adjuvant or neoadjuvant treatment with targeted agents may prove to be particularly beneficial for specific patient groups, for example, those at a higher risk of relapse after nephrectomy or those with locally advanced but non-metastatic disease. As such, targeted agent use in the neoadjuvant and adjuvant settings merits investigation.

Neoadjuvant treatment may allow tumour downstaging and reduce the tumour size to allow surgery in previously unresectable tumours. However, with the currently available targeted agents, downsizing of the primary tumour is limited, with a median reduction in the longest tumour diameter (in the plane of measurement) of 10–14% and RECIST defined primary partial response rates between 3–5% [42,43]. Ongoing neoadjuvant studies include patients who may become candidates for nephron sparing surgery following treatment [44]. One such example is provided by the ongoing phase II study examining the efficacy of pazopanib when administered prior to surgery in patients with large tumours requiring a radical nephrectomy (NCT01158521; Table 4) [45].

Table 4. Planned and ongoing studies investigating the use of targeted agents in the neoadjuvant and adjuvant setting [45–50]]
Study N PhaseTreatment armsPrimary endpoint
Study evaluating pazopanib given before surgery in patients with kidney cancer requiring radical nephrectomy (NCT01158521)30IIPazopanib once daily for up to 18 weeksRate of partial nephrectomy in patients who would otherwise have required radical nephrectomy
Partial or radical nephrectomy performed ≥7 days after finishing pazopanib
S-TRAC: study comparing sunitinib and placebo for the treatment of patients at high risk of recurrent RCC after surgery (NCT00375674)720IIISunitinib 50 mg (Schedule 4/2) for 1 yearDisease-free survival
Placebo (Schedule 4/2) for 1 year
ASSURE: study comparing sunitinib, sorafenib or placebo in patients with kidney cancer that has been removed by surgery (NCT00326898)1923IIISunitinib for nine treatment cycles (6-weekly cycles)Disease-free survival
Sorafenib for nine treatment cycles
Placebo for nine treatment cycles
SORCE: study comparing sorafenib and placebo in patients with high- and intermediate-risk resected kidney cancer (NCT00492258)1656IIIPlacebo for 3 yearsDisease-free survival
Sorafenib for 1 year and placebo for 2 years
Sorafenib for 3 years
EVEREST: study comparing everolimus and placebo in patients with kidney cancer who have undergone surgery (NCT01120249)1218IIIEverolimusRecurrence-free survival
PROTECT: study comparing pazopanib and placebo in patients with kidney cancer who have undergone surgery (NCT01235962)1500IIIPazopanib for 12 monthsDisease-free survival
Placebo for 12 months

The majority of the planned and ongoing studies in non-metastatic RCC are designed to investigate the use of targeted agents in the adjuvant, postoperative setting in the context of surgical treatment (Table 4). These studies are examining the benefit of adjuvant treatment with a targeted agent following nephrectomy in an effort to determine if adjuvant treatment improves disease-free survival in a selected group of patients. Examples include the S-TRAC (NCT00375674), ASSURE (NCT00326898) and SORCE (NCT00492258) studies [46–48]. The S-TRAC study assesses disease-free survival in high-risk patients (according to the UISS staging system) receiving a nephrectomy prior to randomization to either sunitinib or placebo treatment for 1 year. In the ASSURE study, patients with stage II–IV disease are stratified and randomized to treatment with sunitinib, sorafenib or placebo for a total of nine treatment cycles following nephrectomy. The SORCE study includes patients with high- and intermediate-risk resected RCC, receiving a nephrectomy prior to randomization to one of the three study groups: sorafenib treatment for 3 years, sorafenib treatment for 1 year followed by placebo for 2 years or placebo treatment for 3 years. Other examples of studies examining the utility of adjuvant treatment include the EVEREST study (NCT01120249) [49] which examines the efficacy of everolimus for preventing disease recurrence and the PROTECT trial (NCT01235962) [50] evaluating the efficacy and safety of pazopanib versus placebo for 1 year in patients with localized or locally advanced RCC following nephrectomy.

An additional consideration for administering targeted agents in the context of surgical treatment is the associated safety profile of the selected agent. Potential concerns include delayed wound healing, tumour regrowth in the off-treatment period or progression of disease during treatment interruption prior to surgery. Factors such as the elimination half-life are also important considerations when selecting a targeted therapy for administration prior to surgery. The elimination half-lives of sunitinib and its primary active metabolite range between 40–60 h and 80–110 h, respectively [51]. For sorafenib, the elimination half-life is between 25–48 h while bevacizumab has an elimination half-life of 18 days in a typical female patient and 20 days for a typical male patient [52,53]. Pazopanib is eliminated with mean half-life of 30.9 h following administration of the recommended 800 mg dose [54]. As such, it is necessary to determine which agents may provide the greatest benefit from treatment prior to nephrectomy.

Data from small studies highlight the potential differences between the available agents and may help to guide treatment selection in this setting. For example, in a study examining sunitinib administration prior to nephrectomy, delayed wound healing (Clavien classification of surgical complications, Grade I) was observed in five of 52 patients [55]. In the short period off treatment in this study (ranging between 1–14 days), disease progression occurred in 25% of patients although a large proportion (71%) experienced disease stabilization on re-introduction of therapy [55]. In addition, results suggested that it may be safe to continue sunitinib treatment until 1 day prior to surgery and patients re-started sunitinib therapy a median of 21 days following surgery [55]. In another phase II study, bevacizumab treatment was stopped 4 weeks prior to surgery. However, 20.9% of patients experienced delayed wound healing 4 weeks postoperatively [56].


Treatment selection should also include a consideration of additional factors. An understanding of the molecular mechanisms underlying tumorigenesis can serve to guide treatment selection. The identification of both predictive and prognostic markers may serve a dual purpose, allowing identification of patients most likely to benefit from a selected treatment as well as the assessment of response to treatment to determine efficacy of the selected therapeutic agent [57]. Further considerations also include patient stratification into risk categories, based on clinical and non-clinical characteristics, in order to select treatment. In this context, the tolerability profile of the agent also becomes a consideration. However, it is worth noting that treatment associated side effects are often manageable with the application of appropriate therapeutic management strategies, indicating that efficacy remains a primary consideration for treatment selection.


Recent advances in unravelling the angiogenic pathway have led to significantly improved outcomes for the treatment of clear cell mRCC, the most commonly occurring histological subtype. Similar therapeutic advances in the management of less common subtypes (papillary [15%], chromophobe [∼5%] and collecting duct [<1%]) have not occurred, possibly due to the lack of comparable breakthroughs in the knowledge of the underlying pathogenic pathways [58]. The majority of trials with targeted agents excluded patients with non-clear cell RCC. However, the phase III temsirolimus trial included patients with tumours of any histology [12]. Approximately 20% of patients enrolled in the study had non-clear cell RCC. A subanalysis of the trial indicated that efficacy of temsirolimus was similar among patients with clear cell and other histologies [59]. A total of 59% of patients with clear cell mRCC and 68% of those with other histologies receiving temsirolimus experienced tumour reductions [59]. Additionally, both the sunitinib and sorafenib expanded-access trials included patients with non-clear cell histology (588 and 202 patients, respectively) [60,61]. Both trials demonstrated evidence of clinical activity with these agents in patients with non-clear cell RCC [60,61]. However, this requires further assessment, and several ongoing trials are investigating the efficacy of targeted agents in non-clear cell RCC. For example, the phase II ASPEN trial (NCT01108445) [62] is comparing everolimus and sunitinib for the treatment of non-clear cell mRCC. Furthermore, it is hoped that increased understanding of tumour biology, coupled with the identification of both prognostic and predictive markers, may allow therapy to be tailored to optimize individual patient outcomes through the selection of the most optimal treatment for each patient and augment therapy management. In addition, biomarkers may also allow outcomes from treatment to be predicted.

Several tumour biomarkers may have a future role as prognostic indicators for occurrence, recurrence, progression and/or survival in patients with advanced RCC. VEGF polymorphisms have been identified as being associated with an increased risk of developing RCC [63]. Variations in microRNA-related genes are associated with survival and recurrence in patients with RCC [64].

Biomarkers serving as predictors of response to treatment have also been identified by several studies. A recent prospective study enrolled previously untreated patients to receive sunitinib [65]. The efficacy and tolerability of sunitinib in 16 key single nucleotide polymorphisms (SNPs) in nine genes were examined. A total of 95 patients were included in the analysis of toxicity and 89 patients in the analysis of efficacy [65]. Two polymorphisms in VEGFR3 were identified as indicators of reduced efficacy with sunitinib while polymorphisms in CYP3A5*1 may predict toxicity in patients receiving sunitinib treatment [65]. Several other biomarkers such as p53, CAIX, CD147 and tumour pAKT have also been identified as potential predictors of response [66–68]. Similarly, retrospective analyses conducted for patients treated with pazopanib have also identified genetic markers for efficacy in these patients [69,70]. A multiplatform analysis of 129 plasma samples from pazopanib-treated patients carried out to identify cytokines and angiogenic factors (CAFs) associated with response and tumour burden indicated that lower concentrations of hepatocyte growth factor (HGF), IL-6 and IL-8 were associated with greater tumour shrinkage. Low IL-6, HGF and E-Selectin concentrations were associated with prolonged PFS and plasma IL-6 and IL-8 correlated with greater tumour burden [70].

In addition, the occurrence of a particular side effect from treatment may also serve as a predictive marker for efficacy or response to treatment. In a retrospective analysis of RCC patients treated with axitinib, occurrence of hypertension appeared to correlate with outcome [71]. Similarly, in a retrospective analysis of patients treated with sunitinib, hypertension was associated with improved clinical outcomes, indicating that it may be a viable biomarker for antitumour efficacy in these patients [72]. The incidence of hypothyroidism and the association with survival also remains under investigation. A prospective analysis of patients treated with sunitinib or sorafenib demonstrated a significant correlation between the occurrence of subclinical hypothyroidism and the median duration of survival, indicating that hypothyroidism may serve as a predictive marker for treatment outcomes in mRCC [73].

However, the utility of biomarkers identified by retrospective studies and their application to clinical practice warrants further investigation in prospective clinical trials. This is particularly important in light of the potential benefits that biomarker analyses could bring to maximizing treatment outcomes.


Patient profiling, including staging of RCC and assigning patients into risk categories, is an important concept as it allows prediction of tumour behaviour and therefore patient prognosis, prediction of disease recurrence and patient survival, prediction of response to therapy and minimization of treatment exposure, reducing the incidence of unnecessary toxicity and maximising patient outcome [74].

The Memorial Sloan-Kettering Cancer Centre (MSKCC) criteria identified haemoglobin (less than the lower limit of normal [LLN]), lactate dehydrogenase (LDH, greater than 1.5 times the upper limit of normal [ULN]), corrected calcium (greater than the ULN), Karnofsky performance status (less than 80%) and interval from diagnosis to treatment (less than 1 year) as the most significant risk factors for predicting survival in RCC patients treated with IFN-α as first-line systemic therapy [75]. These were used to create a risk model whereby patients were stratified into one of three risk groups based on the number of prognostic factors present: favourable risk, intermediate risk or poor risk. Although these prognostic factors were initially derived based on immunotherapy as first-line treatment, it has now been shown that these same prognostic factors were identified in a trial of sunitinib as first line therapy [76].

In addition, a retrospective study conducted by Heng et al. [77] to validate the MSKCC criteria identified two clinical and four laboratory values as prognostic factors for survival. These included haemoglobin less than the LLN, corrected serum calcium greater than the ULN, Karnofsky performance status less than 80% and interval from diagnosis to treatment of less than 1 year as well as absolute neutrophil counts greater than the ULN and platelets greater than the ULN. These factors were also used to stratify patients into favourable, intermediate and poor risk categories according to the number of prognostic factors present.

Although the analyses described above were conducted retrospectively and require further validation, they provide valuable clinically relevant information and illustrate the importance of patient stratification to assist with the selection of treatment for patients with mRCC.

In solid tumours, the evaluation of patient response can be either an earlier event, response rate, or longer term events such as PFS and OS. RECIST criteria have traditionally been used to evaluate response rate and have generally been correlated with the later events. RECIST criteria are based on the change in one-dimensional measures of the sum of longest diameters (also known as the baseline sum diameters and include the longest diameter for non-nodal lesions and the short axis for nodal lesions) of tumour target lesions measured through imaging. A partial response is defined as a 30% reduction in the sum of longest diameters [78]. In mRCC studies, with anti-angiogenic agents, these measurements may not accurately reflect treatment effects. Therefore, correlations between response rate and PFS/OS are not as consistent. Recently, several studies have suggested that either a smaller reduction in the sum of longest diameters of 10% [79] or combining a measure of tumour reduction with an evaluation of tumour blood flow via CT scanning [80] may provide more significant correlations to PFS. Choi critieria, which define partial response as ≥10% decrease in tumour size or ≥15% decrease in tumour attenuation on CT scan, may also be an alternative method for response evaluation in mRCC. A recent study showed that Choi criteria were significantly better than RECIST criteria for predicting PFS and OS [81].

Several other imaging techniques are also under investigation for evaluating response to antinagiogenic agents. These include dynamic contrast-enhanced magnetic resonance imaging or ultrasound (DCE-MRI and DCE-US, respectively) [82,83]. For example, in a study evaluating early response to sunitinib treatment in mRCC patients, robust correlations were observed between functional DCE-US parameters and classic assessments, including disease-free and overall survival [83]. Novel forms of MRI such as arterial spin labelling MRI or blood oxygen level-dependent (BOLD) imaging [84] are also under investigation in order to assess tumour vascular responses to antiangiogenic therapies.


Targeted agents have a distinct safety and tolerability profile, related partially to their mechanisms of action. Key adverse events associated with tyrosine kinase inhibitors include fatigue, skin toxicities, gastrointestinal symptoms, stomatitis and hypertension [85]. Similarly, the mTOR inhibitors temsirolimus and everolimus are themselves associated with their own particular patterns of toxicity, including metabolic abnormalities (such as hyperglycaemia and hypercholesterolaemia), fatigue and rash [85]. The majority of adverse events are mild-to-moderate in severity. However, some patients will experience higher grade adverse events, which will require effective management. A systematic review of adverse events associated with sorafenib, sunitinib and temsirolimus in phase I, II and II trials showed an overall incidence range of adverse events of <1% to 72%. The incidence of grade 3 or 4 adverse events was lower, occurring in <1 to 13% of patients for sorafenib, <1 to 16% for sunitinib and 1 to 20% for temsirolimus [86]. The incidence of grade 3/4 adverse events with pazopanib has been shown to range from <1 to 5% [54]. Importantly, many adverse events are manageable with appropriate strategies which include prevention, recognition and care (Table 5) [86–88]. Application of practical strategies to manage adverse events is important for maintaining patients on treatment for long periods, at the most optimal dose, thereby increasing the potential for achieving maximal clinical outcome from treatment. However, in some cases, dose reduction or treatment interruption may be necessary [86–88].

Table 5. Therapy management strategies for the management of adverse events associated with targeted agent treatment [86–88]]
Adverse eventManagement strategies
Fatigue and asthenia Pretreatment considerations:
• Educate patients regarding possible occurrence and provide psychological support
• Encourage modification of daily activities to conserve energy
• Recommend moderate physical exercise to help reduce levels of fatigue
During treatment:
• Assess for underlying factors: hypothyroidism, anaemia, depression, emotional distress, sleep disturbance
• Treat underlying factors according to standard treatment
• Provide supportive counselling as required; educating patients on viewing fatigue in the wider context of benefit from treatment
• Dose modifications are infrequently or rarely required
Thyroid dysfunction Pretreatment considerations:
• Monitor for baseline thyroid function and treat according to standard practice
During treatment:
• Monitor for thyroid hormone levels on a regular basis
• Treat according to standard medical practice when clinical signs or biological disturbances occur
Skin disorders Pre-treatment considerations:
• Educate patients using visual illustrations and leaflets and brochures regarding possible occurrence
• Advise patients on possible depigmentation of hair and skin
• Conduct full foot examination and consult with podiatrist – treat any pre-existing hyperkeratosis prior to commencing therapy
• Advise patients to reduce pressure on affected areas
During treatment:
• Advise patients to wear thick soled shoes
• Recommend shock absorbers and hydrocolloidal bandages (grade 2–3 hand-foot syndrome)
• Podiatrist and dermatologist consultations as required for supportive treatment
• Use emollient creams and topical treatments containing urea or salicyclic acid. Topical creams with corticosteroids may be used for patients experiencing painful erythema
• Treatment interruption or dose reduction may be required if hand-foot syndrome grade ≥2 affects patients until resolution to grade 0 or 1
Oral changes, stomatitis and mucositis Pretreatment considerations:
• Educate patients on possible occurrence and symptoms
• Consult dieticians for modification of diet
• Switch to paediatric toothpaste and soft toothbrush
• Advise patients to avoid alcohol
During treatment:
• Advise patients to use bicarbonate mouthwashes containing paracetamol with morphine or codeine to help symptom control
• Tetracain-hydrochloride gels, camomile, sage, arnica and zinc may also be beneficial
• Dose delay or reduction may be required to reduce grade 3 or 4 mucositis to grade ≤1
Gastrointestinal toxicity Pretreatment considerations:
• Educate patients about possible symptoms
• Advise patients to reduce intake or discontinue use of stool softeners, fibre supplements or laxatives to help prevent diarrhoea
• Consultation with a dietician may be beneficial for patients
• Nausea and vomiting may be prevented by a low dose of a single agent such as corticosteroid
During treatment:
• Antiemetic medication may be used as secondary prophylaxis
• Bulking agents or anti-diarrhoeal medication may be used
• Severe diarrhoea may be managed by treatment interruption until severity decreases
• Dose reduction may be used at recurrence of grade 3 diarrhoea and should be employed at recurrence of grade 4 diarrhoea
Hypertension and cardiovascular events Pretreatment considerations:
• Conduct a full cardiovascular assessment before starting treatment
• Monitor for hypertension and treat according to standard medical practice
• Stabilise blood pressure prior to initiating treatment
During treatment:
• Monitor cardiovascular status, including blood pressure, regularly during treatment
• Monitor for signs and symptoms of congestive heart failure or changes in ejection fraction, particularly in patients with cardiac risk factors
• Treat with antihypertensive therapy according to normal management protocols
• Dose adjustments or interruptions may be necessary until hypertension is controlled or if the left ventricular ejection fraction is <50% and >20% below baseline
• Treatment may need to be discontinued in patients with symptoms of congestive heart failure
Haematological toxicity Pretreatment considerations:
• Advise patients on possible occurrence, good diet and eliminate hypothyroidism as a cause
During treatment:
• Monitor blood counts regularly
• Manage and treat according to local guidelines
• Dose delay or reduction may be required in patients with low neutrophil and granulocyte counts


With improvements in the prognosis for patients with mRCC, attributable in a large part to the application of targeted agents for the treatment of these patients, focus is shifting towards maximizing clinical benefit from targeted therapies. As such, factors other than efficacy data are being considered when selecting a treatment strategy, particularly with a view towards ensuring that the most appropriate treatment is selected for each patient.

An understanding of resistance mechanisms, which limit clinical efficacy and long-term outcomes for patients with mRCC, is contributing towards the rational selection of targeted agents in first and subsequent lines of therapy. Sequential therapy is now used routinely in clinical practice and offers the potential to extend patient survival through the use of the available targeted agents in sequence. Data available from retrospective analyses as well as the phase III everolimus trial (RECORD-1) indicate that it is possible to treat patients with both a VEGF inhibitor or an mTOR inhibitor, following prior treatment with a VEGF inhibitor. This indicates that sequential therapy is likely to become the mainstay of mRCC treatment, particularly in view of the potential benefits offered by the several new agents currently under development.

Other aspects of treatment for either primary mRCC or those with non-metastatic disease but high risk of recurrence are under investigation and include the role and timing of treatment with targeted agents around surgery and the associated benefits. The results of ongoing studies in the presurgical, neoadjuvant and adjuvant settings are likely to provide further clarification regarding this issue and to guide further the optimal use of targeted agents. It is possible that, based on the results of such studies, the treatment algorithm for RCC will evolve to differentiate further between the targeted agents, for example, highlighting those agents which may be used to downstage or downsize a tumour or those that may provide benefit in patients at high risk of recurrence or relapse.

The progress in the understanding of the molecular biology of RCC is driving the identification of biomarkers associated with the disease and response to treatment. The potential benefit offered by prognostic and predictive markers warrants further investigation as it may allow treatment to be tailored to optimize outcomes for individual patients and maximize clinical benefit. Another aspect of treatment selection remaining under consideration is the stratification of patients into risk groups as this plays a crucial role in the selection of treatment based on clinical and non-clinical factors. Finally, the importance of therapy management is also clear. The implementation of practical strategies to educate patients, pre-empt the occurrence of some side effects and to manage others once they occur, all play a crucial role in maintaining patients on treatment at the most optimal dose for as long as possible in order that maximal clinical benefit is gained by the patient.

In conclusion, the treatment of advanced RCC is evolving with greater understanding of the factors contributing to successful individually tailored treatment of patients with this difficult to treat disease, involving a multidisciplinary approach. Results of ongoing and planned clinical trials, coupled with advances in the understanding of the underlying biology of RCC, are likely to contribute towards even greater improvements in the prognosis for patients with mRCC in the near future.


Medical writing support was provided by Minal Kotecha at ACUMED (Tytherington, UK) and was funded by Pfizer Inc.


Axel Bex received honoraria for taking part in advisory boards for Pfizer, GSK and Bayer. In line with institutional policy the honoraria have not been received personally but transferred to a research account. Martin Gore is a member of the advisory board/ speaker bureau for Pfizer, Roche, Astellas, Bayer, BMS, Astra Zeneca, GSK and Novartis.