Signal-transduction inhibitors in renal cell carcinoma


  • Nicholas J. Vogelzang,

    1. Nevada Cancer Institute, University of Nevada School of Medicine, Las Vegas, NV, USA, and Department of Medical Oncology, San Camillo Forlanini Hospital, Rome, Italy
    Search for more papers by this author
  • Cora N. Sternberg

    Corresponding author
    1. Nevada Cancer Institute, University of Nevada School of Medicine, Las Vegas, NV, USA, and Department of Medical Oncology, San Camillo Forlanini Hospital, Rome, Italy
    Search for more papers by this author

Cora N. Sternberg, Department of Medical Oncology, San Camillo Forlanini Hospital, Rome, Italy.




vascular endothelial growth factor (receptor)


platelet-derived growth factor (receptor)


von Hippel-Lindau


hypoxia inducible factor


(receptor) tyrosine kinase


phosphatidyl inositol-3-kinase


protein kinase C


mitogen-activated protein kinase


USA Food and Drug Administration


European Agency for the Evaluation of Medical Products


progression-free survival


Eastern Cooperative Oncology Group


time to progression


hormonal therapy




RCC comprises ≈ 3% of adult malignancies and 90–95% of neoplasms arising from the kidney. Immunotherapy with interferon (IFN) and interleukin-2 has produced responses in ≈ 15% of patients, with only a few selected patients benefiting in overall survival. Results with chemotherapy in the treatment of RCC have been equally disappointing.

This review focuses on the recent extremely promising results obtained with orally available small-molecule kinase signal-transduction inhibitors [1–4]. Vascular endothelial growth factor (VEGF) antibodies and combinations including anti-VEGF are reviewed in another article in this issue. The purpose is to appraise and categorize some of the knowledge about these new molecules and offer a glimpse towards future strategies in the treatment of RCC.


RCC is specifically suited for targeting with inhibitors of several kinases that are relevant to tumour proliferation and angiogenesis, including VEGF, raf, platelet-derived growth factor (PDGF) and its receptor (PDGFR). Sporadic RCC is frequently characterized by loss of the von Hippel–Lindau (VHL) tumour suppressor gene (mapped to the short arm of chromosome 3). This results in an increased concentration of hypoxia inducible factor-1 (HIF-1) The resulting overexpression of HIF-1, in turn, leads to the induction of hundreds of genes and proteins that affect many signalling pathways that, in part, assist the cell to respond to hypoxia and which stimulate angiogenesis. These proteins include glucose transporter proteins, erythropoietin, and growth factors and their receptors, e.g. VEGF, PDGF and TGF-α[3].

Signal transduction in cancer cells is a complex process that involves receptor tyrosine kinases (RTKs) that in turn trigger cytoplasmic kinases. There are many cellular signalling pathways that work independently, in parallel, and/or through interconnections, to promote cancer development, growth and survival. The three major signalling pathways in cancer cells include the phosphatidyl inositol-3-kinase (PI3K)/AKT, protein kinase C (PKC), and mitogen-activated protein kinase (MAPK)/Ras-Raf-MEF-ERK signalling cascades. Cancer cells show ‘oncogene addiction’, in that they depend for survival on one pathway. In clear cell RCC that pathway seems to be the VEGF/PDGK/PI3K/AKT/mTOR/S6/HIF pathway, as clinical trials have shown that highly selective or specific blocking of the kinases involved in this signalling pathway has been associated with major responses and a survival advantage.

The Ras-Raf-MEK-ERK pathway is activated by EGFR signalling but blockade of the EGF/MAPK/ras pathway with agents such as gefitinib or erlotinib has shown only sporadic responses in RCC. Blockade of the PKC pathway has not yet been studied in RCC. While cancer cells can show ‘oncogene addiction’, normal stromal or host cells do not. For example, VEGF is produced by tumour cells as well as stromal cells, and ligand binding by VEGFR stimulates endothelial cell proliferation [5]. The Ras-Raf-MEK-ERK pathway also promotes endothelial cell proliferation. The Ras-Raf-MEK-ERK pathway might be activated by a variety of upstream signals, e.g. activating mutations in Ras and Raf [6,7], as well as TK activity from receptors localized to the plasma membrane. This pathway is linked to angiogenesis and includes a role for EGFR, VEGF [8] and PDGF [9]. Inhibition of the Raf pathway inhibits VEGF-mediated endothelial cell angiogenesis [10]. Improved understanding of the complexity of signal-transduction processes and their roles in cancer and the stromal cells (i.e. endothelial cells and pericytes) supporting the malignancy has suggested that simultaneous inhibition of several key kinases at the level of receptors and/or downstream serine/threonine kinases might optimize the overall therapeutic benefit. This suggests that optimal therapeutic approaches to RCC could involve targeting several molecules found in both the tumour and supportive tissues. Several drugs that inhibit Ras, Raf or MEK are under clinical investigation (AZD 6244, ISIS 5132, CI-1040), although results are too premature to be reported [11,12].

Numerous multi-targeted TK inhibitors that target the VEGF/AKT pathway have been developed. Two agents have already been approved by the USA Food and Drug Administration (FDA) and the European Agency for the Evaluation of Medical Products (EMEA) for use as oral single-agent therapy in metastatic RCC, i.e. sunitinib (Sutent®, Pfizer) and sorafenib (Nexavar®, Bayer) and it is likely that other signal-transduction inhibitors in various stages of development, will also be active and approved. Characteristics of novel agents and TK inhibitors are illustrated in Figure 1.

Figure 1.

Kinome dendrogram.

Preclinical data suggest that treatment with an EGFR TK inhibitor such as erlotinib (Tarceva, OSI Pharmaceuticals, Inc., USA) or gefitinib (Iressa, AstraZeneca) in the presence of VEGF blockade (bevacizumab) might result in synergistic action. Moreover, VEGF blockade might be critical in preventing resistance to EGFR inhibition, and EGFR blockade might limit the circulating levels of VEGF and thereby allow more complete VEGF blockade [12].

Considering that many biological pathways are implicated in tumour proliferation, antitumour activity should theoretically be increased by using multi-targeted agents, or by combining different agents to allow vertical or horizontal inhibition of crucial pathways. Table 1 shows some of the new agents and the pathways that are known to be inhibited by these agents.

Table 1.  Molecular-targeted therapies for RCC: target inhibition
Sorafenib XX X X 
Sunitinib XX   X 
Erlotinib/cetuximab/ lapatinib   X   X
Axitunib XX     
Tersimolimus/ Everolimus/Ariad     X  
Pazopanib XX   X 

Kinase inhibitor specificity can be evaluated by measuring the binding of small molecules to the ATP site of kinases. Specificity can vary substantially even among compounds that target the same kinase The kinase dendrograms shown in Fig. 1 show the differences among some of the compounds in clinical development [13].


Sunitinib is an oral multi-targeted RTK inhibitor with antitumour and anti-angiogenic activity through targeting of PDGFR-β, VEGFR-2, KIT and Flt3 receptors. Sunitinib has shown antitumour activity by inhibiting RTKs expressed by cancer cells directly involved in cancer proliferation and survival, and RTKs expressed on endothelial or stromal cells (pericytes) that support cancer growth [14,15].

In two independent phase II single-arm multicentre trials, sunitinib was given at a dose of 50 mg/day to patients with metastatic RCC, who had failed cytokine therapy with interleukin-2 or IFN. Patients with a good or intermediate prognosis received treatment in 6-week cycles, with 4 weeks on and 2 weeks off therapy. Sunitinib gave response rates of 42% in cytokine-refractory RCC [16,17]. The response was evaluated by an independent third-party imaging laboratory and by treating physicians (investigator assessment), for efficacy analyses. An objective response was confirmed in 36 of 105 (34%) evaluable patients according to an independent third-party assessment, with a median (95% CI) progression-free survival (PFS) of 8.3  (7.8–14.5) months. The most common adverse events reported by patients were fatigue in 30 (28%) and diarrhoea 21 (20%). Neutropenia, elevation of lipase, and anaemia were the most common laboratory abnormalities, in 45 (42%), 30 (28%) and 27 (26%) patients, respectively [17].

In a groundbreaking randomized phase III trial, sunitinib was compared with IFN in patients who had not previously received cytokines [18]. Untreated patients with clear cell RCC were randomized 1 : 1 to receive sunitinib (6-week cycles: 50 mg orally once daily for 4 weeks, followed by 2 weeks off) or IFN-α (6-week cycles: s.c. injection 9 MU given three times weekly). A trial of 690 patients was designed to have 90% power to detect a 35% improvement in median PFS from 20–27 weeks (4.6–6.2 months; two-sided unstratified log-rank test; significance level 0.05). The primary endpoint was PFS. To account for patient withdrawal, ineligibility, etc., the trial was designed to have 750 patients randomized 375 to sunitinib and 375 to IFN-α. Results of a planned analysis on the primary endpoint, PFS, were presented.

An independent central review assessed that the median (95% CI) PFS was 11 (10–12) months for sunitinib, vs 5 (4–6 months) for IFN-α (hazard ratio 0.415; 95% CI 0.320–0.539; P < 0.001). The objective response rate was 31% for sunitinib vs 6% for IFN-α (P < 0.001); 632 patients (85%) remained alive, with 49 deaths on the sunitinib arm and 65 on the IFN-α arm. In all, 8% withdrew from the study due to adverse events on the sunitinib arm, vs 13% on the IFN-α arm. Survival results were premature and are eagerly awaited.

Even before presenting these results, FDA regulatory approval for advanced RCC was granted in January 2006 and EMEA approval followed shortly thereafter.

The side-effects of sunitinib include asthenia, hand-foot syndrome, hypertension, and diarrhoea. Haematological abnormalities are also frequent, leading some to consider that a better dosage schedule might be 37.5 mg administered in a continuous-dosing regimen [19].


Sorafenib is an oral agent that was designed as a c-and b-raf kinase inhibitor. The Ras/Raf signalling pathway is a mediator of tumour cell proliferation and angiogenesis. Sorafenib has also been found to inhibit several RTKs, among them VEGFR-2, PDGFR-β, Flt-3 and c-KIT [20].

A phase II randomized discontinuation trial reported results in 202 patients with metastatic RCC [21]. The trial was initially planned for solid tumours, particularly colon cancer, with known raf mutations, but the cohort of RCC patients was expanded because there were positive results. The effects of sorafenib on VEGFR and PDGFR were subsequently assessed. Patients were evaluated after a lead-in phase of 12 weeks. If there was ≥ 25% tumour shrinkage then the drug was continued. Those with disease stabilization were randomized to receive sorafenib or placebo for another 12 weeks; half were progression-free at 24 weeks, vs 18% of patients randomized to placebo (P = 0.008). PFS was 24 and 6 weeks, respectively, for patients treated with sorafenib or placebo (P = 0.009).

These results led to a phase III randomized trial, known as TARGET (Treatment Approaches in RCC Global Evaluation Trial) [22]; 905 patients with clear cell histology, measurable disease, having failed one previous systemic therapy in the last 8 months, with good or intermediate prognosis, and an Eastern Cooperative Oncology Group (ECOG) performance status of 0–1 were eligible. The primary aim of the trial was to assess overall survival and the secondary endpoint was PFS. In addition to tumour control in 80%, sorafenib gave a significantly longer PFS than placebo. The median PFS was 5.5 months, vs 2.8 months for placebo (hazard ratio 0.44; 95% CI 0.35–0.55; P < 0.01). The median overall survival was 19.3 months for sorafenib vs 15.9 months for placebo (hazard ratio 0.77; 95% CI 0.63–0.95; P = 0.015), a trend that was not as significant as that specified for the interim analysis. In an update, Eisen et al.[23] 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.

Doppler ultrasonography with perfusion software (vascular recognition imaging) was used to predict drug efficacy. A decrease in tumour vascularization correlated with the response on CT at 6 weeks. This noninvasive imaging technique is under investigation [24].

The TARGET trial is the largest randomized study in advanced RCC. It is positive in its primary endpoint in a planned preliminary analysis. The follow-up continues but the toxicity is considered manageable. The FDA granted approval for this drug for the second-line therapy of RCC in 2005; approval was also granted by EMEA. No data are available on poor-risk patients.


Tumour angiogenesis is also stimulated by growth factors through the PI3K-AKT-mTOR signal-transduction pathway. The mTOR/AKT pathway is a nutrient-sensing pathway and in a feedback loop, is up-regulated by HIF [25]. VHL mutations predict for sensitivity to mTOR inhibitors in the laboratory [26]. A well-known mTOR inhibitor is rapamycin (sirolimus; Rapamune), an immunosuppressive agent. Recently, rapamycin analogues such as temsirolimus (CCI-779, TEMSR), everolimus (RAD001) and Ariad (AP23573), have been developed [12,27,28].

Temsirolimus was given to 111 heavily pretreated patients with advanced refractory RCC, evaluating doses of 25–250 mg. Despite an objective response rate of only 7%, and minor responses in 26%, the overall tumour growth control was 70% and the time to progression (TTP) was 6 months. The median overall survival was 15.0 months [29].

A phase I dose-escalation study of temsirolimus combined with IFN-α in advanced patients who had received no more than two previous systemic therapies was also reported. Of 71 patients, 96% had had a previous nephrectomy and 55% previous immunotherapy. The maximum tolerated dose was 15 mg of temsirolimus weekly combined with 6 mIU of IFN-α s.c. three times weekly. Dose-limiting toxicities were fatigue, stomatitis, and nausea and vomiting. There were eight (11%) partial responses and 21 (30%) patients with stable disease. The median TTP was 9.1 months [30].

A phase III multicentre trial was performed at 209 sites in 26 countries in previously untreated patients with poor-prognosis metastatic RCC. Temsirolimus as a single agent vs combined therapy with IFN-α vs IFN-α alone was studied and presented in 2006 [31]. Pretreatment features associated with shorter survival after cytokine therapy were a low Karnofsky performance status (<80%), high lactate dehydrogenase level (>1.5 × normal), low haemoglobin level, high serum calcium, and the absence of nephrectomy [32]. In that study, poor-risk patients with advanced RCC and no previous systemic therapy were defined as having three or more of six risk factors (the five Motzer criteria and more than one metastatic disease site); 67% had had a previous nephrectomy.

Patients were randomized among three arms: single-agent IFN-α up to 18 MU s.c. three times a week, single agent temsirolimus 25 mg i.v. once per week and a combination of temsirolimus 15 mg i.v. once per week plus IFN-α 6 MU s.c. three times a week. The primary study endpoint was overall survival. Of 626 patients enrolled, 442 died and the follow-up was 13 months after the last patient was enrolled. In this second planned interim analysis, patients treated with temsirolimus had statistically longer survival than those treated with IFN-α. The median survival with temsirolimus was 10.9 (8.6–12.7) months, vs 7.3 (6.1–8.9) months (hazard ratio 0.73; 95% CI 0.57–0.92, P = 0.007) on IFN-α. Survival on the combined arm was 8.4 (6.6–10.2) months, and not statistically different (hazard ratio 0.95; 95% CI 0.76–1.2, P = 0.691) from IFN-α.

Temsirolimus, as a single agent (25 mg i.v. weekly), significantly improved overall survival and PFS in patients with metastatic RCC and poor-risk features, compared with IFN-α. There was a 3.6-month (49%) improvement in median overall survival and a 1.8-month (95%) improvement in median PFS.

The four most common adverse events were asthenia, anaemia, skin rash and dyspnoea. This is the first study to show a statistically significant improvement in survival in patients with advanced poor-risk RCC.

Everolimus (RAD001) an orally bioavailable mTOR inhibitor, was given to 28 patients at a dose of 10 mg/day and induced a response in seven of 25 patients. Additional patients had minor responses [33,34]. Importantly, mTor inhibitors act synergistically with angiogenesis inhibitors, and phase I trials of several combinations are underway, e.g. bevacizumab plus everolimus [35] and valatinib plus temsirolimus.


Pazopanib is another novel multi-targeted inhibitor of VEGFR-1, -2, -3, PDGFR-α and β, and c-kit, that is being evaluated in RCC [36]. Two global studies are being conducted in patients with advanced or metastatic RCC. A phase II randomized discontinuation design of open-label drug will accrue 160–230 patients. Patients with objective responses at the end of 12 weeks in the lead-in phase will continue on open-label pazopanib. Patients who have progressive disease will be withdrawn from study. Patients with stable disease after 12 weeks will be randomized to either pazopanib or placebo. These patients will be followed for another 16 weeks and assessed for progressive disease at the final analysis.

A multicentre international phase III study was designed to evaluate pazopanib in patients with locally advanced and/or metastatic RCC who either have failed first-line cytokine-based therapy or have been intolerant to such therapy [36]. Given the rapidly changing landscape, a study amendment permits patients to be entered who have not received previous cytokines, using stratification between first- and second-line therapy. The primary endpoint is PFS, and the principal secondary endpoint is overall survival. About 350 eligible patients will be enrolled and randomized in a 2 : 1 ratio to receive either 800 mg pazopanib once daily or placebo. The study was initiated in April 2006 and has completed enrollment in March 2007.


Axitunib (AG-013736) is an orally available, potent inhibitor of VEGFR-1, -2, -3 and PDGFR, now manufactured by Pfizer. In a single-arm, multicentre, phase 2 study, patients with cytokine-refractory RCC were treated with axitunib 5 mg/kg twice daily. After a median follow-up of 1 year, there was a partial response in 46% of the patients; overall, 86% of patients showed some type of tumour reduction [37]. The median time to disease progression has not been achieved after 12 months. Treatment-related adverse events were similar to the side-effects of sunitunib and included mild to moderate hypertension (controlled by standard antihypertensive therapy), diarrhoea, fatigue, nausea, proteinuria, changes in hair colour, and hoarseness. These results show that axitunib is an active agent in RCC and is worth further development. It is being studied in sorafenib-resistant patients and in a variety of other diseases, including colorectal cancer.


The angiogenesis inhibitor valatinib (PTK 787/ZK 222584, PTK/ZK) blocks all known VEGFR TKs, including the lymphangiogenic VEGFR-3, and has shown promising activity in metastatic RCC in a phase I trial [38], but has not yet been studied in a phase II trial. In a phase III trial in colorectal cancer it did not improve overall survival when added to a standard chemotherapy regimen.


Lapatinib is an orally active, reversible inhibitor of EGFR and ErbB2 TKs [39]. A randomized open-label phase III trial of oral lapatinib vs hormonal therapy (HT) in patients with advanced RCC who express EGFR and/or ErbB2 by immunohistochemistry (IHC) was reported in 2006 [40]. The main endpoints were TTP and overall survival. Patients with advanced RCC of any histology who had failed first-line cytokine therapy were stratified by performance status and number of metastatic sites. The median TTP was 15.3 weeks for lapatinib vs 15.4 weeks for HT, and the median survival was 46.9 weeks for lapatinib vs 43.1 weeks for HT. In a subset analysis of a subgroup of 241 patients with EGFR-overexpressed disease (3+ by IHC), the median TTP was 15.1 weeks for lapatinib vs 10.9 weeks for HT, and the median overall survival was 46.0 weeks for lapatinib vs 37.9 weeks for HT. Given the problems in interpreting IHC, these results (not confirmed by fluorescence in situ hybridization) and only positive for 3+ EGFR expression, are not very promising as a single agent in RCC. Results with this drug in combination with capecitabine in breast cancer are of greater interest [41].


In 2006, several phase I trials were reported of sorafenib and bevacizumab, a so-called ‘vertical blockade’ of the VEGF pathway. No trials were reported of sorafenib plus sunitunib or of sunitunib plus bevacizumab. Sosman et al.[42] reported that neither drug could be given at the recommended phase II dose, but that there were responses in eight of the 18 patients with RCC. Data of the synergistic toxicity and efficacy of sorafenib plus bevacizumab were also presented [43]. Responses were encouraging in patients with RCC or ovarian cancer, but there was also significant hypertension and proteinuria, suggesting synergistic toxicity.


Patients with recurrent disease after nephrectomy have micrometastatic disease at the time of surgery. Depending upon stage and grade, there is a 35–65% recurrence rate in patients with locally aggressive tumours. Adjuvant therapy trials using cytokines and vaccines have failed to show a survival advantage.

The use of effective therapy might reduce the risk of relapse, and thus adjuvant trials have been initiated. The Assure trial (1332 patients) is an intergroup study sponsored by ECOG. After nephrectomy patients are stratified by The UCLA Integrated Staging System stage (II–V) and histological subtype (clear cell or not) among three arms, to 1 year of adjuvant sunitinib, sorafenib or placebo. The primary endpoint is disease-free survival.

In Europe the MRC will lead the SOURCE trial; after nephrectomy, 1420 patients with high- and intermediate-risk RCC will be randomized to 3 years of sorafenib, 1 year of sorafenib and 2 years of placebo, or 3 years of placebo. The primary endpoint is metastasis-free survival. Phase II neoadjuvant trials of sunitinib (Cleveland Clinic) and RAD-001 (UCLA) have also been initiated.


Table 2 lists some of the agents, their various phases of development and the pathways that are known to be inhibited. Varying response rates and toxicity profiles among TKIs suggest that there are differences among these agents. Are these agents cross-resistant? Is one agent better than another? It is thought that resistance to one signal-transduction inhibitor or monoclonal antibody might be mediated by certain pathways, and that tumours could still be sensitive to inhibition of different pathways by another TKI or monoclonal antibody.

Table 2.  The development of signal-transduction inhibitors in RCC
ProductCompanyStatus (trial phase)Pathways inhibited
  1. FGFR, fibroblast growth factor receptor.

SunitunibPfizerIIIVEGFR, PDGFR, c-Kit, Flt-3
LapatinibGlaxoIIIEGFR and HER-2
PazopanibGlaxoIIIVEGFR & PDGFR, c-Kit
VandetanibAstraZenecaIIVEGFR & EGFR
CEP-7055Cephalon/SanofiIVEGFR, Flt-3
ZD2171AstraZenecaIVEGFR, PDGFR
LBQ258NovartisIVEGFR, FGFR, PDGFR, Lck, Fyn
AMG-706AmgenIVEGFR, PDGFR, c-Kit
AEE-788NovartisIEGFR, HER-2, VEGFR
Indolyl quinolinoneMerckI 

Several trials are evaluating these agents in sequence. Sunitinib has shown substantial activity in bevacizumab-refractory RCC [44]. Second-line trials in patients failing sunitinib and sorafenib are ongoing. Other trials are evaluating axitunib in sorafenib-refractory RCC, and varying doses of sorafenib in patients who have received anti-VEGF therapy. Combinations of signal-transduction inhibitors and combination with cytokines are also of interest. Are the studied doses and schedules optimal? Trials are evaluating a continuous oral dosage schedule of sunitinib (phase III renal EFFECT trial, 499 patients). Can combined therapy improve the outcome? The ECOG randomized phase II trial (360 patients) will evaluate bevacizumab vs bevacizumab + temsirolimus vs sorafenib vs temsirolimus + sorafenib.

There are many questions about RCC in this era of biologically targeted therapy. When can therapy be stopped? Is there a ‘rebound’ phenomenon? What are the long-term effects of these agents? Will chronic therapy imply chronic side-effects? Temsirolimus is active in high-risk patients; are the others also active? What is the effect on the primary tumour? Can response rate be used in combination with surgery? Combinations should clearly be explored with upstream targets such as HIF and mTOR. How much do we really know about their sequential use? Perhaps the timing or order in which these drugs are given might not be crucial, provided that patients have time to respond.

Direct comparisons between the various compounds are needed. Combinations with interleukin-2 might also be of interest. Are these agents immunosuppressive? Are cytokines of any use in the second-line setting? Are there prognostic and predictive factors? PFS and TTP are unproven surrogates for survival; do we need better surrogate endpoints to evaluate new agents, since the RECIST and WHO response data have limited usefulness. Predictive markers in tissue and blood, and novel imaging methods such as dynamic imaging, might help to identify those patients who are benefiting from therapy. We do not know very much about the role of these agents in other than clear cell RCC and studies are underway in the adjuvant setting. Quality of life is crucial in evaluating the benefits of these agents. They all have important side-effects and should only be administered by physicians who have experience.


Targeted therapies have shown robust activity in metastatic RCC. The VEGF signal-transduction pathway is clearly an important target for therapy in the treatment of advanced RCC. There is increasing awareness that validating therapeutic targets is necessary for the discovery of new drugs, and to verify their success. There is a strong rationale for targeting many pathways, particularly angiogenesis pathways in RCC. No direct comparisons of these agents have been made and thus all have emerged as promising and viable options for patients with metastatic RCC. Clinical trials of these agents in the adjuvant setting and in combination are ongoing to optimize timing, sequence and combination of therapies in RCC.

Molecular profiling is heralding the future of prognosis, staging and treatment. Further exploration of biological targets, angiogenesis inhibition, and EGFR antagonists are providing new possibilities. Efforts to improve results include an international effort to identify prognostic factors [45]. The next few years should be characterized by new rational treatment strategies based on inhibition of specific biological pathways.


None declared. Source of funding: Nicholas J. Vogelzang received research support from Bayer & Pfizer.