Skip to main content
PMC home page
Therapeutic Advances in Medical Oncology logoLink to Therapeutic Advances in Medical Oncology
. 2013 Mar;5(2):119–131. doi: 10.1177/1758834012463624

Imaging techniques as predictive and prognostic biomarkers in renal cell carcinoma

Paul Nathan 1,✉, Anup Vinayan 2
PMCID: PMC3556873  PMID: 23450112

Abstract

A number of imaging modalities are showing promise as predictive and prognostic biomarkers in advanced renal cell carcinoma. This review discusses progress to date in this exciting area and identifies areas of future promise.

Keywords: biomarkers, imaging, prediction, prognosis, renal cell carcinoma

Introduction

Therapeutic options for patients with advanced renal cell carcinoma (RCC) have dramatically improved over the last few years with the advent of a number of novel agents targeting both tumour vasculature and tumour growth. Sunitinib [Motzer et al. 2007b], bevacizumab in combination with interferon [Escudier et al. 2007b], temsirolimus [Motzer et al. 2007a], sorafenib and everolimus [Motzer et al. 2008] have all become standard agents for the treatment of this disease.

There is a clinical need to identify which patients are likely to derive most and least benefit from targeted therapies early in the course of treatment, that is, to predict response, and to identify which patients have aggressive or indolent disease, that is, to determine prognosis.

RCC is highly vascular, driven at least in part by the frequent overexpression of vascular endothelial growth factor (VEGF) [Nicol et al. 1997]. As all of the new therapeutic agents appear to have at least some effect upon tumour vasculature, imaging modalities that provide a quantitative assessment of the change in vascularity may have a particular role in this disease.

This review discusses imaging modalities that have shown promise as either predictive or prognostic tests in advanced RCC and identifies areas of future opportunity.

Predictive biomarkers

Validated tools, which could predict clinical benefit from a therapy, would lead to significant improvements in clinical management. Early identification of responders, nonresponders or subpopulations of patients at risk for side effects would help to identify patients who have greatest or least benefit from a specific intervention. Predictive biomarkers would therefore help to avoid exposing patients to the risk of unnecessary side effects from treatments from which they would be destined to derive little or no benefit.

There are now a significant number of new therapies which have proven to be clinically effective, but which have not succeeded in passing health–economic assessments by reimbursement authorities. The cost-effectiveness analyses that bodies such as the National Institute for Health and Clinical Excellence perform would be significantly impacted upon if it were possible to identify subpopulations of patients who would derive most benefit from the drugs. This would lead to a greater possibility that new therapies would be available for patients once efficacy had been demonstrated.

Imaging-defined response evaluation criteria in solid tumours (RECIST) [Therasse et al. 2000] have been the basis of treatment response assessment in oncology since their introduction. The criteria were developed and validated to assess responses to conventional therapies such as chemotherapy and radiotherapy rather than the new generation of targeted agents. RECIST criteria are based upon change in size of target lesions. However, antivascular agents appear to be frequently clinically active without inducing the significant size change required for a RECIST-defined response. No imaging modality has yet been validated that can predict the treatment response of RCC treated with targeted therapies.

Primary RCC and its metastases are noted to be highly vascular and are associated with an upregulation of VEGF. Most sporadic and familial clear cell RCCs have an inactive VHL gene product either because of bi-allelic alterations in the gene or post-transcriptional modification. Lack of functional VHL gene product results in stabilization of hypoxia inducible factor (HIF) 1, 2 and increased transcription of factors such as VEGF and platelet-derived growth factor. Activation of multiple pathways driven by these and other factors results in increased angiogenesis in these tumours. This is thought to underpin the significant clinical activity of antiangiogenic targeted therapies in RCC. Imaging techniques that can assess vascularity and changes in vascularity of these tumours may therefore have potential to be useful as predictive biomarkers.

Assessment of changes in vascularity

The gold standard for assessing vascularity and angiogenesis in a tumour is by measuring the mean vascular density, which is performed by counting vascular structures directly with histology. This technique has limitations as it is invasive and cannot be performed repeatedly to assess treatment response in humans. Also, as it does not explore the whole treatment volume, there is an increased chance of sampling error due to tumour heterogeneity [Lebret et al. 2007; Renshaw et al. 1997].

A number of clinically applicable functional imaging techniques are capable of quantitatively assessing the vascularity of tumours in vivo pre and post treatment. These techniques include dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), dynamic contrast-enhanced computed tomography (DCE-CT), dynamic contrast-enhanced ultrasound (DCE-US), diffusion-weighted MRI (DW-MRI), arterial spin label MRI (ASL-MRI) and positron emission tomography (PET) with oxygen-labelled water. Each of these techniques provides quantitative or semiquantitative data related to blood flow and some can also provide information on blood volume, cellularity or vessel permeability. DCE imaging techniques such as DCE-MRI, DCE-CT and DCE-US use exogenous contrast agents to differentiate vascular characteristics, while other MRI techniques as ASL-MRI and DW-MRI use the inherent changes in magnetic resonance between tissues without the use of an external contrast agent. The latter group has a potential advantage as 20–30% of patients with RCC have impaired renal function, which increasingly precludes the use of exogenous contrast agents.

Dynamic contrast-enhanced imaging techniques

Dynamic functional imaging using exogenous contrast media centres on quantitative analysis of biodistribution of contrast media in tissues. These techniques use the difference in pharmacokinetics of the contrast agent between tissues and tumour to provide tumour-specific and tissue-specific information [Marcus et al. 2009].

DCE-MRI as a biomarker of angiogenesis has been tested and validated over the last decade in a variety of tumours. The gold standard correlation was with immunohistochemical microvessel density measurements. Investigators have attempted to correlate imaging with tissue expression of proangiogenic growth factors including VEGF (broad correlations in some studies and no correlations in others) [Padhani and Dzik-Jurasz, 2004; Schlemmer et al. 2004].

The relationship between contrast concentration and signal intensity is complex and nonlinear in DCE-MRI [Kiessling et al. 2007]. Along with the concentration, it also depends on the imaging sequence used, machine setup, inherent heterogeneity of the tumour and native relaxation rate of the tissues [Padhani, 2003].

The rate of delivery and washout of contrast media into the tumour vasculature bed depends on several physiological characteristics of the tumour vascular microenvironment. Depending on the techniques used, parameters such as Ktrans (transfer constant, related to the influx of contrast agents from plasma to the tumour interstitial space), Kep (efflux of agents back to the vasculature), Ve (effective volume of distribution), rBF (relative blood flow), rBV (relative blood volume), IAUGC (initial area under gadolinium curve) can be obtained. The values are then fitted with statistical models for analysis.

In 2008, Flaherty and colleagues published a pilot study, which used DCE-MRI assessment in patients with metastatic RCC treated with sorafenib. This study showed that inhibition of tumour vascular permeability by sorafenib, demonstrated by a reduction in Ktrans, was associated with improved progression free survival (PFS) (p = 0.01). Elevated baseline Ktrans was a statistically significant predictive marker for favourable response to treatment (p < 0.02) [Flaherty et al. 2008]. A prospective randomized phase II study again with sorafenib in RCC by Stadler and colleagues also showed a good correlation of high baseline Ktrans with improved PFS (p < 0.027). In this study, however, change in Ktrans or IAUGC90 were not significantly related to PFS [Hahn et al. 2008]. Further trials are underway to evaluate the use of DCE-MRI in RCC.

Most modern MRI scanners have the ability to perform these image acquisitions but there are several limitations to widespread use. Imaging protocols need to be optimized according to the area to be scanned and the type of tumour being analysed. The appropriate software needs to be available and there also remain issues regarding quantification, tumour heterogeneity and motion artefact. Low molecular weight gadolinium, which is used as standard, has a limitation as these small agents rapidly leak out of the highly permeable vasculature in the tumour. This in turn affects the data acquisition in dynamic imaging as this is defined by vascular permeability and tumour perfusion. Gadolinium in higher molecular weight conjugates could be used to overcome this problem, as these are less sensitive to vascular permeability.

Lamuraglia and colleagues used DCE-US along with a new perfusion software as a predictive tool in metastatic RCC [Lamuraglia et al. 2006]. Thirty patients who were enrolled in a randomized clinical trial comparing sorafenib with placebo were recruited. DCE-US examination was performed at baseline, week 3 and week 6. The good responders were defined by a combination of a decrease in contrast uptake greater than 10% and a stable or decrease in tumour size. The responders identified at week 3 had a statistically significant improvement in PFS (p < 0.0004) and overall survival (OS) (p < 0.0004) over the nonresponders.

A recent prospective multicentre study by Lassau and colleagues investigated the use of DCE-US as a predictive biomarker in patients treated with antiangiogenic agents. A total of 539 patients with different tumour types including 157 patients with RCC were recruited from 19 centres in France. Perfusion parameters were collected from these patients with DCE-US at baseline, on day 7, day 14, day 30, day 60 and 2 monthly thereafter. This was correlated to the RECIST assessment performed 2 monthly. They showed that a decrease in area under the curve (AUC) by more than 40% at 1 month was predictive of response for time to progression (TTP) (p < 0.01) and OS (p < 0.04) [Lassau et al. 2012].

A simultaneous study with DCE-US from Toronto by Williams and colleagues in 17 patients with metastatic RCC treated with sunitinib showed a decrease in median tumour fractional blood volume by 73.2% after 2 weeks of treatment with a disruption-replenishment infusion method of microbubble and a significant decrease in bolus peak with the bolus method. In this study no definite correlation was noted with best response as assessed by RECIST [Williams et al. 2011].

US is a widely available and inexpensive technique. Therefore performing frequent and serial examinations at short intervals is possible when using this technique. However, the main problem with US is the limited ability to penetrate through large areas of tissue when trying to image deeply located tumours, and more specifically, the inability to penetrate through air or bone. Because RCC frequently metastasizes to the lung and mediastinum, the use of DCE-US in large cohorts of patients is unknown. A potential advantage of DCE-US is the fact that the microbubbles are intravascular and therefore, in theory, the quantification of tumour perfusion should not be affected by vessel permeability. However, the technique tends to be more operator dependent and those with the relevant skills are not available in all locations.

To date, no studies have been published using DCE-CT as a tool for treatment response prediction in RCC. This technique does hold promise as the concentration kinetics of the contrast are related directly and linearly to enhancement and analysis is therefore relatively straight forward [Miles, 2002]. However, there is little general agreement about the techniques to use for modelling and further quantitative analysis. Evidence from rectal carcinoma showed that baseline blood flow and blood volume were significantly lower in nonresponders compared with responders treated with chemoradiotherapy [Bellomi et al. 2007]. The study also showed a significant decrease in blood flow, blood volume and permeability surface area product post treatment. CT scanning is the commonest tool used at present to monitor treatment response, therefore it would be a practical and easy option to incorporate DCE-CT protocols in almost all centres once a general agreement is obtained and the technique is validated further.

Imaging techniques that use no external contrast agent

A number of MRI techniques which do not resort to an external contrast agent have been developed and are being assessed as biomarkers of treatment response. Between 20% and 30% of patients with RCC are unable to have external contrast due to renal impairment and techniques that do not resort to external contrast are thus valuable[Nathan et al. 2010]. These techniques use the magnetic field to modify the magnetic properties of the tissue including the tumour.

ASL-MRI uses the inherent nuclear spin of endogenous water protons to label arterial blood. Quantitative images of blood flow to the tissue can be generated by differential labelling inside and outside the tissue of interest. The technique has potential to be used as a quantitative marker for response assessment to antivascular treatment. The feasibility of this technique to monitor response to antiangiogenic therapy in RCC metastases was demonstrated by de Bazelaire and colleagues [de Bazelaire et al. 2005]. In a phase II trial of 10 patients, ASL-MRI and DCE-MRI were used to assess the treatment response in patients with metastatic RCC treated with an antiangiogenic agent (PTK787/ZK222584). The study showed that changes in blood flow detected by ASL-MRI at 1 month correlated with the change in tumour size measured at 4 months or at disease progression (p < 0.01) [de Bazelaire et al. 2008]. If validated in larger studies, ASL-MRI thus might prove to be an important biomarker in early assessment of treatment response.

Techniques such as blood oxygen level dependent MRI (BOLD-MRI) and DW-MRI which use no external contrast are other methods with potential to predict treatment response. BOLD-MRI uses the difference in magnetic properties of oxyhaemoglobin and deoxyhaemoglobin and hence BOLD signals are related to tumour hypoxia. DW-MRI evaluates the property of water diffusion depending on the Brownian motion of the water molecules in tissues. Desar and colleagues noted that sunitinib-induced antiangiogenic effects in RCC can be reliably measured with DW-MRI as early as 3–10 days after starting the treatment [Desar et al. 2011]. This needs further evaluation before it can be reliably used as a predictive biomarker. There are as yet no published data with either of these techniques to predict responses in RCC to targeted therapies.

Assessment of change in metabolic activity

Fluorodeoxyglucose (FDG)-PET/CT scan is a valuable tool, which has already established its role in the assessment and management of many tumour types. By combining a short-lived radioactive tracer with a biologically active molecule, PET scan provides a functional map of the disease investigated. The most common biological molecule used in oncology is FDG. This glucose analogue is rapidly and preferentially taken up by the rapidly growing malignant cells, thus leading to intense radiolabelling of that tissue. A combined CT and PET scan provides a combination of metabolic activity with anatomical localization of tissues. The use of FDG-PET/CT in treatment response assessment, that is, as a predictive marker in RCC, looks promising but has not been fully evaluated. A recent study by Powles and colleagues looked at the use of sequential FDG-PET scanning as a correlative marker of OS in patients with RCC treated with sunitinib [Powles et al. 2010]. A total of 44 patients were enrolled in this phase II study. Patients had three FDG-PET scans: before treatment, at 4 and 16 weeks. Reduction in maximum standardized uptake value (SUVmax) by 20% was considered a response and was correlated with OS. The study showed that high baseline SUVmax suggested a trend towards low OS [hazard ratio (HR) 3.30; 95% confidence interval (CI) 1.36–8.45]. Metabolic response was noted on the PET scan after 4 weeks. However, disease progression according to the 16-week scan correlated with a decreased OS (HR −5.96; 95% CI −2.43 to +19.02).

Similar results were noted in smaller trials of 12 patients and 14 patients by Minamimoto and Revheim and colleagues respectively [Minamimoto et al. 2010; Revheim et al. 2011]. Using a 20% reduction in FDG-PET uptake, Minamimoto and colleagues were able to separate good responders with a mean PFS of 233.8 days and poor responders with a PFS of 75 days.

A pharmacodynamic study using serial 3deoxy3’ fluorothymidine PET (FLT-PET) scans on patients at baseline, during treatment with sunitinib and after withdrawal of drug within cycle 1 showed an increase in cellular proliferation during sunitinib withdrawal [Liu et al. 2011]. This is noted to associate with VEGF ligand levels and early exploratory analysis suggests that the extent of flare could predict poorer clinical benefit from the drug.

New criteria for response assessment

RECIST criteria depend upon the change in sum of the unidimensional measurements of target lesions. However, the clinical benefit does not always correlate with change in size alone in tumours treated with targeted therapies. Emerging evidence shows that some patients continue to derive clinical benefit from continuation of sunitinib after documented disease progression based on RECIST criteria [Teo et al. 2012]. This retrospective analysis of 39 patients treated with sunitinib showed that such use of a tyrosine kinase inhibitor (TKI) beyond progression gives prolonged disease control and is safe. The use of RECIST alone as a measure of treatment response is therefore arguable.

Choi criteria

Choi proposed an alternative set of criteria for the assessment of gastrointestinal stromal tumours (GIST) treated with the targeted agent imatinib [Choi, 2008; Choi et al. 2007]. The Choi criteria require either a change in size or a change in density (contrast enhancement) to predict a response. They have been found to be more reliable than RECIST in monitoring GIST responses to treatment using PET-CT as a gold standard. It has been suggested that the Choi criteria could be suitable for assessment of targeted therapies in other cancers, but this has not yet been extensively evaluated.

In renal cancer, van der Veldt and colleagues have analysed the use of Choi criteria in treatment response assessment in RCC and found them to correlate better to TTP than RECIST [van der Veldt et al. 2010]. A total of 55 patients with metastatic RCC treated with sunitinib were analysed retrospectively. The pretreatment and first evaluation post-treatment scan were analysed with RECIST and Choi criteria separately and correlated with PFS and OS. At first evaluation, according to the RECIST criteria, seven patients had a response, 38 had stable disease and 10 had progressive disease. According to the Choi criteria, these numbers were 36, 6 and 13 respectively. In patients having a partial response, the Choi criteria had a significantly better predictive value for PFS and OS (p < 0.001 for both) than RECIST (p = 0.689 and 0.191 respectively). The predictive value of RECIST increased to a statistically significant level only when best response during treatment was taken into account.

The Choi criteria and RECIST were again compared with a smaller cohort of 22 patients treated with sorafenib reported by Hittinger and colleagues [Hittinger et al. 2011]. Even though the Choi criteria defined more patients as responders, they did not differentiate between patients who did and those who did not have therapeutic improvements in PFS or OS.

’Modified’ Choi criteria

We proposed a modification of the Choi criteria in the assessment of treatment response to targeted therapy in metastatic RCC in which both a change in size and density are required [Nathan et al. 2010]. The standard Choi criteria require a decrease in size by 10% or a decrease in CT contrast enhancement by 15% for a response. We suggested a modification of the Choi criteria in which both a minimum 10% size decrease and a 15% reduction in enhancement are required to define a response. We performed a retrospective study of 32 patients with metastatic RCC treated with TKIs. Twelve patients were excluded as they did not have contrast-enhanced scans due to renal dysfunction. Both baseline and 12-week post-treatment contrast-enhanced scans were collected and analysed. According to the response obtained on the 12-week scan, patients were assigned to progressive disease, stable disease or partial response using RECIST, Choi and modified Choi criteria, correlated with TTP. Patients who had a partial response according to the modified Choi criteria had a statistically significant longer TTP than those who were deemed to have stable disease (mean 448 versus 89 days, p = 0.002). Neither RECIST nor Choi criteria defined partial responders went on to have a significant improvement in TTP compared with those who had stable disease. Further larger trials are planned to validate these results. CT enhancement was measured in the arterial phase rather than in the more usual portovenous phase in this study. This may be relevant as renal tumours are highly vascular and enhance well in the arterial phase [Lee et al. 2005; Raptopoulos et al. 2001].

Size and attenuation CT criteria/morphology, attenuation, size and structure criteria

Another study exploring the use of a combination of change in size and CT attenuation to predict treatment response in metastatic RCC was published by Smith and colleagues [Smith et al. 2010a]. Contrast-enhanced CT scans of 53 patients with metastatic RCC treated with either sunitinib or sorafenib were analysed. In calculating the attenuation data, Smith and colleagues used volumetric mean attenuation of target lesions in place of mean attenuation in the most representative axial image. They proposed new imaging criteria, size and attenuation CT (SACT) criteria. The criteria defined favourable response as any of the following: a decrease in tumour size of at least 20%; a decrease in tumour size of at least 10% and at least half of nonlung target lesions with decreased mean attenuation of at least 20 Hounsefield units (HU); or one or more lung lesions with decreased mean attenuation by at least 40 HU. Unfavourable response was defined as any of the following: an increase in tumour size of at least 20%; new metastasis; marked central fill in of a target lesion; or new enhancement in a homogenously hypoattenuating nonenhancing mass. A comparison was made between the SACT criteria, RECIST criteria and a modification of the Choi criteria (with volumetric attenuation data of the whole tumour rather than mean attenuation in an axial slice) in predicting a PFS greater than 250 days in this patient population. The favourable response defined by the SACT criteria had a sensitivity of 75% and a specificity of 100% in identifying patients with PFS greater than 250 days. These values were 16% and 100%, and 93% and 44% for the RECIST criteria and volumetric modified Choi criteria, respectively.

The same group attempted to improve on the SACT criteria by including specific morphologic or structural changes (e.g. necrosis) and eliminating the need for volumetric three-dimensional analysis in their new criteria [referred to as morphology, attenuation, size and structure (MASS) criteria] [Smith et al. 2010b]. According to the MASS criteria, favourable response was defined as no new lesion and any of the following: a decrease in tumour size of at least 20%; one or more predominantly solid enhancing lesions with marked central necrosis or marked decreased attenuation (≥40 HU). An unfavourable response was defined as an increase in tumour size of at least 20% in the absence of marked necrosis or marked decreased attenuation; and new metastasis, marked central fill-in, or new increased enhancement. Other changes, which did not fit in with these criteria were classified as an indeterminate response. The MASS criteria showed an increased sensitivity of 86% and specificity of 100% in recognizing good responders (TTP and improved disease specific survival p < 0.0001).

The group further recommended that the combination of memorial sloan kettering cancer centre (MSKCC) criteria and MASS criteria together could give a high overall accuracy in identifying patients with PFS less than 1 year or at least 1 year [Smith et al. 2011].

Tumour burden and growth rate

Basappa and colleagues investigated the effect of tumour burden (TB) with treatment with VEGF-targeted therapy. Sixty-nine patients were retrospectively analysed. Median TB and site of metastasis were examined. With multivariate analysis, at baseline, the total number of metastases (<10) and the TB above the diaphragm (<6.5 cm) were independent positive predictors of OS [Basappa et al. 2011]. TB (p = 0.003) and total metastases (≤12) were noted as predictors of OS at TTP.

Initial tumour size and the rate of size reduction in response to targeted therapy were predictive markers reported by Yuasa and colleagues[Yuasa et al. 2011]. In this 139-patient retrospective analysis, a linear, moderate to strong association was noted between the pretreatment tumour size and reduction in growth rate on therapy (correlation coefficient −0.441, p < 0.001). When a threshold value of 23.9 mm was applied, it was noted that the smaller tumours showed a better rate of shrinkage (p < 0.001).

Use of tumour growth rate (TGR) was retrospectively analysed by Ferte and colleagues [Ferte et al. 2012]. Images from patients recruited to large phase III trials with sorafenib (TARGET) and everolimus (RECORD) were assessed (n = 902). TGR was defined as the value obtained by dividing tumour shrinkage by time between two evaluations. All treatment periods, including before, under treatment (at first cycle), at progression and after treatment interruption, were noted. The results showed that the TGR under treatment was significantly decreased with both sorafenib and everolimus. High TGR under treatment was associated with poor PFS (HR 2.6) and OS (HR 2.3) in patients treated with sorafenib and poor OS (HR 1.2) in the smaller everolimus cohort. TGR after interruption was significantly higher in both sorafenib and everolimus cohorts than TGR at progression (14.6 versus 31 and 17.9 versus 32.1, respectively). The inclusion of TGR thus allows the practitioner to evaluate the tumour response more carefully.

The threshold of tumour response as change in sum of longest diameter (ΔSLD) to mammalian target of rapamycin inhibitor everolimus was investigated by Oudard and colleagues [Oudard et al. 2012]. The data from the phase III RECORD-1 trial were analysed and a series of arbitrary thresholds were identified to attempt to distinguish responders from nonresponders. With response defined by the optimal threshold of −5%ΔSLD, the median PFS was 8.4 months for the responders and 5.0 months for the nonresponders (HR 2.4).

Prognostic biomarkers

Imaging biomarkers that provide an insight into the natural course of the disease are of potential use in identifying patients who are most or least likely to benefit from treatment as well as in interpreting clinical trials. The prognosis of any malignancy could be related to a variety of factors. RCC is a heterogeneous group of diseases. There are a number of histological subtypes and within each subtype there is a widely varying natural history between patients. Some tumours are very aggressive while others remaining indolent. Some patients experience stable metastatic disease for a significant period of time that does not require systemic treatment until progression is seen.

Several clinicopathological prognostic models are designed to assess survival in the postnephrectomy [Leibovich score, University of California, Los Angeles (UCLA) score] [Leibovich et al. 2005; Zisman et al. 2001] and metastatic setting (Motzer score, UCLA score) [Motzer et al. 1999]. These prognostic models use histological features, including Fuhrman grade and necrosis, performance status and a variety of biochemical markers for categorization into prognostic groups.

The MSKCC (Motzer) criteria classified patients into good, intermediate and poor prognostic groups depending on five factors: primary tumour remaining in situ, Eastern Cooperative Oncology Group performance status, anaemia, lactate dehydrogenase level and hypercalcaemia. Median survival in patients with zero risk factors was 20 months but median survival in the intermediate-risk group (one or two risk factors) and the poor prognostic group (more than two risk factors) were 10 months and 4 months respectively [Motzer et al. 1999]. The MSKCC criteria have been widely used in many phase III trials of RCC evaluating antiangiogenic agents either to stratify patients in trial analysis [Escudier et al. 2007b; Motzer et al. 2007b, 2008] or to define entry criteria in other trials [Escudier et al. 2007a; Hudes et al. 2007].

Evidence from these trials showed that the beneficial effects of these drugs were maintained across prognostic groups but were significantly greater in good and intermediate prognostic groups. Hence prognostic grouping of RCC has an important role in selection of treatment and also in determining the time of introduction of systemic treatment. Imaging techniques can potentially act as noninvasive tools in identifying these prognostic factors. These include assessment of histological subtype, nuclear grading and in the accurate staging of disease.

Assessment of histological subtypes

There has been debate as to whether prognosis of RCC varies significantly with histological subtype. Studies have in part been hampered by small numbers of the rarer subtypes and associated lack of control for other prognostic factors. RCC can be subclassified according to histology as clear cell, papillary type I and II, chromophobe, sarcomatoid, collecting duct and a variety of very rare and unclassified subtypes(World Health Organization classification) [Ebele et al. 2004; Kovacs et al. 1997]. The clear cell (conventional) subtype is the most common and accounts for approximately 70% of cases. The papillary subtype accounts for around 15–20% of cases. The type II papillary and other subtypes are less common. There is good evidence that tumours with sarcomatoid elements have a poorer prognosis than the majority of tumours without [Cangiano et al. 1999].

With the emerging possibility that treatment decisions may be influenced by histological subtype, together with the fact that core biopsy is associated with diagnostic inaccuracy [Lebret et al. 2007] a noninvasive technique to assess histological subtype would be of potential clinical value. A number of different imaging modalities have been assessed for this purpose. These include multislice CT, DW-MRI and DCE-MRI.

Sheir and colleagues attempted to use multislice CT to differentiate histological subtypes. They reviewed 87 CT images of patients with RCC retrospectively and compared them with histopathology. Biphasic CT scans of the kidneys were performed using unenhanced, corticomedullary phase (CMP) and excretory phase (EP) (30 and 300 s after contrast injection respectively) protocols. It was noted that the degree of enhancement in the CMP and EP was the most valuable parameter in attempting to differentiate between histological subtypes (p < 0.001). Mean CT attenuation (+2 standard deviations) in the CMP was 138.2 ± 38.0 HU for clear cell, 89.2 ± 31.4 HU for papillary and 55.17 ± 24.0 HU for chromophobe tumours. In the excretory phase, clear cell tumours showed an enhancement of 73.0 ± 17.6 HU while papillary and chromophobe had an enhancement of 70.0 ± 10.4 and 33.9 ± 12.1 HU respectively. The difference between subtypes in both the CMP and EP was statistically significant (p < 0.001). The cutoff value for highest accuracy for diagnosis of clear cell carcinoma was 83.5 HU for CMP and 64.5 HU for EP. Other factors such as presence or absence of cystic degeneration, tumour vascularity and pattern of enhancement also supplemented this differentiation [Sheir et al. 2005].

Pedrosa and colleagues attempted evaluation of malignant renal tumours with the help of noncontrast-enhanced and contrast-enhanced MRI and correlated this with histology. The study cohort consisted of 76 patients who had both MRI and histological confirmation [Pedrosa et al. 2008]. Renal masses were analysed for size, location, necrosis, haemorrhage, signal intensity and uniformity of signal intensity, type and extent of fat and contrast enhancement, peri-renal fat invasion and renal vein thrombosis. They were then assigned to one of eight different groups defined according to MRI appearance. The results were compared with the histological analysis and the sensitivity and specificity of these MRI groups were calculated to predict the histological diagnosis and nuclear grade. The MRI classification had a sensitivity of 93% and specificity of 75% for the diagnosis of clear cell subtype, using the presence of solid enhancing mass with or without necrosis. Retroperitoneal collaterals, intravoxel fat signal and necrosis all correlated directly with clear cell subtype. MRI features such as hemorrhagic cyst with peripheral enhancing growth in the wall and solid hypo-enhancing homogeneous masses with low SI on T2-weighted images were predictors of the papillary subtype with good sensitivity (80%) and specificity (94%).

Sun and colleagues explored the use of DCE-MRI as a tool in predicting the histopathological subtype of RCC. They compared the difference in enhancement pattern in CMP and nephrographic phase images with histopathology. In both the CMP and nephrographic phase, clear cell tumours showed maximum signal intensity change and papillary tumours the least. Signal intensity change in the CMP was the most effective parameter differentiating clear cell and papillary subtypes with a sensitivity of 93% and a specificity of 96% [Sun et al. 2009]. DCE-MRI was also useful in predicting the nuclear grade of the tumour as reported by Palmowski and colleagues. Twenty-one patients were analysed with DCE-MRI presurgically; tumour perfusion and tissue–blood ratio were noted for the entire tumour and also for the most highly vascularized area in the tumour. Higher grade tumours had significantly higher perfusion values (p < 0.05) in the most vascular area [2.14 ± 0.89 versus 1.40 ± 0.49 ml/g/min) and the entire tumour (1.59 ± 0.44 versus 1.08 ± 0.38 ml/g/min) [Palmowski et al. 2009].

These imaging techniques are of interest, however their use in clinical practice will require much more robust validation and they are unlikely to replace the use of histological diagnosis in the near future.

Staging of renal cell carcinoma

Accurate staging of RCC is an important factor influencing the prognosis. While CT, MRI, PET and bone scans all are used routinely in staging a variety of different cancers, there is debate regarding the usefulness of some of these techniques in RCC.

Assessment of skeletal metastasis

Nuclear medicine (NM) bone scans with Tc99 provide limited anatomical detail and only provide information about tumour deposits where an osteoblastic reaction has occurred. RCC bony metastases are predominantly osteoclastic, resulting in a low sensitivity of conventional bone scans. Staudenherz and colleagues showed that the sensitivity of bone scintigraphy varied from 10% to 60% in RCC [Staudenherz et al. 1999].

MRI has been suggested as a more sensitive tool in the assessment of skeletal metastasis than NM bone scan. A recent study by Sohaib and colleagues compared the efficacy of whole body MRI scan and NM bone scan in the assessment of bony RCC metastasis. This prospective study of 47 patients showed that even though both bone scan and MRI were highly specific (94% versus 97%), the sensitivity of whole body MRI scan was much superior (94%) to bone scan (62%) in the detection of renal metastasis to bone (p = 0.007). MRI also had the advantage of assessment of soft tissue disease simultaneously [Sohaib et al. 2009].

Assessment of visceral disease

For assessment of lymph node and visceral disease, FDG-PET scan has been found to be extremely sensitive in characterizing metastases from a variety of tumour types. However, there is a significant incidence of false-positive results, which is perhaps not surprising given that FDG is not a cancer-specific tracer [Antoch et al. 2003; Schmidt et al. 2006]. Small lesions in the lung and liver may remain FDG occult due to smearing of the signal as a result of organ motion. Use of PET-CT is also inferior to whole body MRI in cancers with a frequent metastatic spread to the bone, liver and central nervous system [Antoch et al. 2003; Schmidt et al. 2005]. Poor uptake of FDG has also been noted in tumour types such as RCC, prostatic malignancy and low-grade soft tissue malignancies [Schmidt et al. 2006].

The value of FDG PET scans in RCC diagnosis is debated with multiple small studies showing sensitivity ranging from 32% to 100% for initial diagnosis and staging of primary RCC [Montravers et al. 2000; Nakatani et al. 2011]. The largest of these series, published by Kang and colleagues, included 66 patients with RCC and showed a sensitivity of 60% [Kang et al. 2004]. Nakatani and colleagues suggested a higher FDG sensitivity for RCC metastasis but this has not yet been validated [Nakatani et al. 2011]. Larger series of studies are required to validate the use of FDG-PET scan in RCC but as the specificity of this technique was higher than conventional imaging in previous studies, its use would be limited at present as a tool to help resolve dilemmas in selected cases with equivocal conventional imaging but high suspicion.

In a retrospective study by Le Moulec and colleagues, 21 patients with metastatic RCC who received FDG-PET/CT study at baseline and after 2 months were evaluated [Le Moulec et al. 2010]. The study noted that patients with a negative PET/CT had a better outcome (overall median survival not reached versus 23 months, = 0.05). Patients with a median value of SUVmax greater than 5.7 had a short PFS and OS (3 months versus 9 months, p = 0.03) and (23.3 months versus 9.1 months). Another observation in the study was that in eight patients the early PET/CT findings were consistent with the later CT results.

A recent as yet unpublished study from Japan by Nakaigawa and colleagues examined whether SUVmax from the pretreatment FDG-PET scan could be used as a prognostic marker for survival [Nakaigawa et al. 2012]. In 67patients enrolled and monitored in this study, the researchers noted that the SUVmax varied widely in RCC tumours from an undetectable level increasing to 16.6 (mean 7.6 ± 3.6) with an increasing trend towards poorer prognosis (p < 0.001, HR −1.289). The results were further analysed with SUVmax results stratified as less than 7.0, between 7.0 and 12 and greater than 12.0. Median OS for patients was 1229 ± 991 days, 446 ± 202 days and 95 ± 43 days respectively.

In a separate study of 35 patients, this group suggested that an FDG-PET scan 1 month after treatment initiation could be used as a predictive marker for therapeutic response to TKIs based on whether the tumour size had increased and SUVmax had decreased by less than 20% or by at least 20% [Ueno et al. 2012].

Assessment of mechanism of drug action and role in drug development

Imaging biomarkers have an increasing role in early phase drug development. The phase I trial end points have now moved away from the traditional end points of maximum tolerated dose and toxicity assessment to also incorporate pharmacodynamic end points which evaluate the mechanism of drug action and also to aid evaluation of the therapeutic window of these targeted agents.

With a view to streamlining the drug development process, the US Food and Drug Administration has approved performing exploratory investigator new drug (IND) studies (termed as phase 0 trials). The purpose of these early studies is to assist in the go versus no-go decision for a product using human models than relying on inconsistent animal data. Functional imaging has an important role in this setting as it can provide evidence of drug activity as well as help in selection of the functional dose of the drug.

Discussion

RCC appears to be more sensitive to treatment with antiangiogenic drugs than many other tumours due to the high vascularity of the tumour and its dependence on the VHL pathway in pathogenesis. With an increasing number of therapeutic agents in development, the need for biomarkers is more critical than ever for analysing efficacy and potentially improving the cost–benefit ratio of treatment. There are now emerging data about how these newer techniques could fit into the paradigm of RCC management, starting from initial diagnosis and staging to therapeutic response assessment. These early data need extensive and robust validation. Functional imaging with DCE-MRI and DCE-CT holds promise as predictive and prognostic biomarkers in treatment response in patients with RCC. DW-MRI can give insight into histological subtypes and in the diagnosis of skeletal metastasis, which is appealing and may have utility in clinical practice. Evidence needs to be generated about the use of other techniques in RCC, such as ASL-MRI and MR spectroscopy.

In conclusion, in an area in which biomarker development is the focus of international research activity, imaging as a biomarker in RCC holds most promise in making an impact upon the clinical management of this disease.

Footnotes

Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest statement: The authors declare no conflict of interest in preparing this article.

Contributor Information

Paul Nathan, Mount Vernon Cancer Centre – Medical Oncology, Rickmansworth Road, Northwood, Middlesex HA6 2RN, UK.

Anup Vinayan, Mount Vernon Cancer Centre – Medical Oncology, Northwood, Middlesex, UK.

References

  1. Antoch G., Vogt F., Freudenberg L., Nazaradeh F., Goehde S., Barkhausen J., et al. (2003) Whole-body dual-modality PET/CT and whole-body MRI for tumor staging in oncology. JAMA 290: 3199–3206 [DOI] [PubMed] [Google Scholar]
  2. Basappa N., Elson P., Golshayan A., Wood L., Garcia J., Dreicer R., et al. (2011) The impact of tumor burden characteristics in patients with metastatic renal cell carcinoma treated with sunitinib. Cancer 117: 1183–1189 [DOI] [PubMed] [Google Scholar]
  3. Bellomi M., Petralia G., Sonzogni A., Zampino M. G., Rocca A. (2007) CT perfusion for the monitoring of neoadjuvant chemotherapy and radiation therapy in rectal carcinoma: initial experience. Radiology 244(2): 486–493 [DOI] [PubMed] [Google Scholar]
  4. Cangiano T., Liao J., Naitoh J., Dorey F., Figlin R., Belldegrun A. (1999) Sarcomatoid renal cell carcinoma: biologic behavior, prognosis, and response to combined surgical resection and immunotherapy. J Clin Oncol 17: 523–528 [DOI] [PubMed] [Google Scholar]
  5. Choi H. (2008) Response evaluation of gastrointestinal stromal tumors. Oncologist 13 (Suppl. 2): 4–7 [DOI] [PubMed] [Google Scholar]
  6. Choi H., Charnsangavej C., Faria S., Macapinlac H., Burgess M., Patel S., et al. (2007) Correlation of computed tomography and positron emission tomography in patients with metastatic gastrointestinal stromal tumor treated at a single institution with imatinib mesylate: proposal of new computed tomography response criteria. J Clin Oncol 25: 1753–1759 [DOI] [PubMed] [Google Scholar]
  7. De Bazelaire C., Alsop D., George D., Pedrosa I., Wang Y., Michaelson M., et al. (2008) Magnetic resonance imaging measured blood flow change after antiangiogenic therapy with PTK787/ZK 222584 correlates with clinical outcome in metastatic renal cell carcinoma. Clin Cancer Res 14: 5548–5554 [DOI] [PubMed] [Google Scholar]
  8. De Bazelaire C., Rofsky N., Duhamel G., Michaelson M., George D., Alsop D. (2005) Arterial spin labeling blood flow magnetic resonance imaging for the characterization of metastatic renal cell carcinoma1. Acad Radiol 12: 347–357 [DOI] [PubMed] [Google Scholar]
  9. Desar I., ter Voert E., Hambrock T., van Asten J., van Spronsen D., Mulders P., et al. (2011) Functional MRI techniques demonstrate early vascular changes in renal cell cancer patients treated with sunitinib: a pilot study. Cancer Imaging 11: 259–265 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ebele J., Sauter G., Epstein J., Sesterhenn I. (eds) (2004) World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of the Urinary System and Male Genital Organs. Lyon: IARC [Google Scholar]
  11. Escudier B., Eisen T., Stadler W., Szczylik C., Oudard S., Siebels M., et al. (2007a) Sorafenib in advanced clear-cell renal-cell carcinoma. N Engl J Med 356: 125–134 [DOI] [PubMed] [Google Scholar]
  12. Escudier B., Pluzanska A., Koralewski P., Ravaud A., Bracarda S., Szczylik C., et al. (2007b) Bevacizumab plus interferon alfa-2a for treatment of metastatic renal cell carcinoma: a randomised, double-blind phase III trial. Lancet 370: 2103–2111 [DOI] [PubMed] [Google Scholar]
  13. Ferte C., Albiges L., Soria J., Loriot Y., Fizazi K., Escudier B. (2012) The use of tumor growth rate (TGR) in evaluating sorafenib and everolimus treatment in mRCC patients: an integrated analysis of the TARGET and RECORD phase III trials data. J Clin Oncol 30(Suppl.): abstract 4540. [Google Scholar]
  14. Flaherty K., Rosen M., Heitjan D., Gallagher M., Schwartz B., Schnall M., et al. (2008) Pilot study of DCE-MRI to predict progression-free survival with sorafenib therapy in renal cell carcinoma. Cancer Biol Ther 7: 496–501 [DOI] [PubMed] [Google Scholar]
  15. Hahn O., Yang C., Medved M., Karczmar G., Kistner E., Karrison T., et al. (2008) Dynamic contrast-enhanced magnetic resonance imaging pharmacodynamic biomarker study of sorafenib in metastatic renal carcinoma. J Clin Oncol 26: 4572–4578 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hittinger M., Staehler M., Schramm N., Übleis C., Becker C., Reiser M., et al. (2011) Course of size and density of metastatic renal cell carcinoma lesions in the early follow-up of molecular targeted therapy. Urol Oncol 23 August (Epub ahead of print). [DOI] [PubMed] [Google Scholar]
  17. Hudes G., Carducci M., Tomczak P., Dutcher J., Figlin R., Kapoor A., et al. (2007) Temsirolimus, interferon alfa, or both for advanced renal-cell carcinoma. N Engl J Med 356: 2271–2281 [DOI] [PubMed] [Google Scholar]
  18. Kang D., White R., Jr, Zuger J., Sasser H., Teigland C. (2004) Clinical use of fluorodeoxyglucose F 18 positron emission tomography for detection of renal cell carcinoma. J Urol 171: 1806–1809 [DOI] [PubMed] [Google Scholar]
  19. Kiessling F., Jugold M., Woenne E., Brix G. (2007) Non-invasive assessment of vessel morphology and function in tumors by magnetic resonance imaging. Eur Radiol 17: 2136–2148 [DOI] [PubMed] [Google Scholar]
  20. Kovacs G., Akhtar M., Beckwith B., Bugert P., Cooper C., Delahunt B., et al. (1997) The Heidelberg classification of renal cell tumours. J Pathol 183: 131–133 [DOI] [PubMed] [Google Scholar]
  21. Lamuraglia M., Escudier B., Chami L., Schwartz B., Leclère J., Roche A., et al. (2006) To predict progression-free survival and overall survival in metastatic renal cancer treated with sorafenib: pilot study using dynamic contrast-enhanced Doppler ultrasound. Eur J Cancer 42: 2472–2479 [DOI] [PubMed] [Google Scholar]
  22. Lassau N., Vilgrain V., Taieb S., Lacroix J., Aziza R., Cuinet M., et al. (2012) Evaluation with DCE-US of antiangiogenic treatments in 539 patients allowing the selection of one surrogate marker correlated to overall survival. J Clin Oncol 30(Suppl.): abstract 4618. [Google Scholar]
  23. Lebret T., Poulain J., Molinie V., Herve J., Denoux Y., Guth A., et al. (2007) Percutaneous core biopsy for renal masses: indications, accuracy and results. J Urol 178: 1184–1188 [DOI] [PubMed] [Google Scholar]
  24. Lee E., Heiken J., Huettner P., Na-Chiangmai W. (2005) Renal cell carcinoma visible only during the corticomedullary phase of enhancement. AJR Am J Roentgenol 184(3 Suppl.): S104–S106 [DOI] [PubMed] [Google Scholar]
  25. Leibovich B., Cheville J., Lohse C., Zincke H., Frank I., Kwon E., et al. (2005) A scoring algorithm to predict survival for patients with metastatic clear cell renal cell carcinoma: a stratification tool for prospective clinical trials. J Urol 174: 1759–1763 [DOI] [PubMed] [Google Scholar]
  26. Le Moulec S., Carassou P., Billemont B., Massard C., Vedrine L., Gontier E., et al. (2010) Prospective evaluation of FDG-PET/CT in patients with advanced RCC treated with antiangiogenic therapies. 2010 Genitourinary Cancers Symposium 5–7 March, San Francisco, ASCO [Google Scholar]
  27. Liu G., Jeraj R., Vanderhoek M., Perlman S., Kolesar J., Harrison M., et al. (2011) Pharmacodynamic study using FLT PET/CT in patients with renal cell cancer and other solid malignancies treated with sunitinib malate. Clin Cancer Res 17: 7634–7644 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Marcus C., Ladam-Marcus V., Cucu C., Bouché O., Lucas L., Hoeffel C. (2009) Imaging techniques to evaluate the response to treatment in oncology: current standards and perspectives. Crit Rev Oncol Hematol 72: 217–238 [DOI] [PubMed] [Google Scholar]
  29. Miles K. (2002) Functional computed tomography in oncology. Eur J Cancer 38: 2079–2084 [DOI] [PubMed] [Google Scholar]
  30. Minamimoto R., Nakaigawa N., Tateishi U., Suzuki A., Shizukuishi K., Kishida T., et al. (2010) Evaluation of response to multikinase inhibitor in metastatic renal cell carcinoma by FDG PET/contrast-enhanced CT. Clin Nucl Med 35: 918–923 [DOI] [PubMed] [Google Scholar]
  31. Montravers F., Grahek D., Kerrou K., Younsi N., Doublet J., Gattegno B., et al. (2000) Evaluation of FDG uptake by renal malignancies (primary tumor or metastases) using a coincidence detection {gamma} camera. J Nucl Med 41: 78–84 [PubMed] [Google Scholar]
  32. Motzer R., Escudier B., Oudard S., Hutson T., Porta C., Bracarda S., et al. (2008) Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial. Lancet 372: 449–456 [DOI] [PubMed] [Google Scholar]
  33. Motzer R., Hudes G., Curti B., McDermott D., Escudier B., Negrier S., et al. (2007a) Phase I/II trial of temsirolimus combined with interferon alfa for advanced renal cell carcinoma. J Clin Oncol 25: 3958–3964 [DOI] [PubMed] [Google Scholar]
  34. Motzer R., Hutson T., Tomczak P., Michaelson M., Bukowski R., Rixe O., et al. (2007b) Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med 356: 115–124 [DOI] [PubMed] [Google Scholar]
  35. Motzer R., Mazumdar M., Bacik J., Berg W., Amsterdam A., Ferrara J. (1999) Survival and prognostic stratification of 670 patients with advanced renal cell carcinoma. J Clin Oncol 17: 2530–2540 [DOI] [PubMed] [Google Scholar]
  36. Nakaigawa N., Yao M., Tateishi U., Minamimoto R., Uemura H., Inoue T., et al. (2012) Impact of maximum standardized uptake value (SUVmax) evaluated by 18-fluoro-2-deoxy-dglucose positron emission tomography/computed tomography (FDG PET/CT) on survival for patients with advanced renal cell carcinoma. J Clin Oncol 30(Suppl.): abstract 4626. [Google Scholar]
  37. Nakatani K., Nakamoto Y., Saga T., Higashi T., Togashi K. (2011) The potential clinical value of FDG-PET for recurrent renal cell carcinoma. Eur J Radiol 79: 29–35 [DOI] [PubMed] [Google Scholar]
  38. Nathan P., Vinayan A., Stott D., Juttla J., Goh V. (2010) CT response assessment combining reduction in both size and arterial phase density correlates with time to progression in metastatic renal cancer patients treated with targeted therapies. Cancer Biol Ther 9: 15–19 [DOI] [PubMed] [Google Scholar]
  39. Nicol D., Hii S., Walsh M., Teh B., Thompson L., Kennett C., et al. (1997) Vascular endothelial growth factor expression is increased in renal cell carcinoma. J Urol 157: 1482–1486 [PubMed] [Google Scholar]
  40. Oudard S., Thiam R., Fournier L., Medioni J., Lamuraglia M., Scotte F., et al. (2012) Optimisation of the tumour response threshold in patients treated with everolimus for metastatic renal cell carcinoma: analysis of response and progression-free survival in the RECORD-1 study. Eur J Cancer 48: 1512–1518 [DOI] [PubMed] [Google Scholar]
  41. Padhani A. (2003) MRI for assessing antivascular cancer treatments. Br J Radiol 76(Suppl. 1): S60–S80 [DOI] [PubMed] [Google Scholar]
  42. Padhani A., Dzik-Jurasz A. (2004) Perfusion MR imaging of extracranial tumor angiogenesis. Top Magn Reson Imaging 15: 41–57 [DOI] [PubMed] [Google Scholar]
  43. Palmowski M., Schifferdecker I., Zwick S., Macher-Goeppinger S., Laue H., Haferkamp A., et al. (2009) Tumor perfusion assessed by dynamic contrast-enhanced MRI correlates to the grading of renal cell carcinoma: initial results. Eur J Radiol 74: e176–e180 [DOI] [PubMed] [Google Scholar]
  44. Pedrosa I., Chou M., Ngo L., Baroni R., Genega E., Galaburda L., et al. (2008) MR classification of renal masses with pathologic correlation. Eur Radiol 18: 365–375 [DOI] [PubMed] [Google Scholar]
  45. Powles T., Kayani I., Blank C., Chowdhury S., Horenblas S., Sarwar N., et al. (2010) The safety and efficacy of sunitinib prior to planned nephrectomy in metastatic clear cell renal cancer. J Clin Oncol 28(15 Suppl.): abstract 4603. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Raptopoulos V., Blake S., Weisinger K., Atkins M., Keogan M., Kruskal J. (2001) Multiphase contrast-enhanced helical CT of liver metastases from renal cell carcinoma. Eur Radiol 11: 2504–2509 [DOI] [PubMed] [Google Scholar]
  47. Renshaw A., Lee K., Madge R., Granter S. (1997) Accuracy of fine needle aspiration in distinguishing subtypes of renal cell carcinoma. Acta Cytol 41: 987–994 [DOI] [PubMed] [Google Scholar]
  48. Revheim M., Winge-Main A., Hagen G., Fjeld J., Fosså S., Lilleby W. (2011) Combined positron emission tomography/computed tomography in sunitinib therapy assessment of patients with metastatic renal cell carcinoma. Clin Oncol 23: 339–343 [DOI] [PubMed] [Google Scholar]
  49. Schlemmer H., Merkle J., Grobholz R., Jaeger T., Michel M., Werner A., et al. (2004) Can pre-operative contrast-enhanced dynamic MR imaging for prostate cancer predict microvessel density in prostatectomy specimens? Eur Radiol 14: 309–317 [DOI] [PubMed] [Google Scholar]
  50. Schmidt G., Baur-Melnyk A., Herzog P., Schmid R., Tiling R., Schmidt M., et al. (2005) High-resolution whole-body magnetic resonance image tumor staging with the use of parallel imaging versus dual-modality positron emission tomography-computed tomography: experience on a 32-channel system. Invest Radiol 40: 743–753 [DOI] [PubMed] [Google Scholar]
  51. Schmidt G., Haug A., Schoenberg S., Reiser M.(2006) Whole-body MRI and PET-CT in the management of cancer patients. Eur Radiol 16: 1216–1225 [DOI] [PubMed] [Google Scholar]
  52. Sheir K., El-Azab M., Mosbah A., El-Baz M., Shaaban A.A. (2005) Differentiation of renal cell carcinoma subtypes by multislice computerized tomography. J Urol 174: 451–455 [DOI] [PubMed] [Google Scholar]
  53. Smith A., Lieber M., Shah S. (2010a) Assessing tumor response and detecting recurrence in metastatic renal cell carcinoma on targeted therapy: importance of size and attenuation on contrast-enhanced CT. AJR Am J Roentgenol 194: 157–165 [DOI] [PubMed] [Google Scholar]
  54. Smith A., Shah S., Rini B., Lieber M., Remer E. (2010b) Morphology, Attenuation, Size, and Structure (MASS)criteria: assessing response and predicting clinical outcome in metastatic renal cell carcinoma on antiangiogenic targeted therapy. Am J Roentgenol 194: 1470–1478 [DOI] [PubMed] [Google Scholar]
  55. Smith A., Shah S., Rini B., Lieber M., Remer E. (2011) Utilizing pre-therapy clinical schema and initial CT changes to predict progression-free survival in patients with metastatic renal cell carcinoma on VEGF-targeted therapy: a preliminary analysis. Urol Oncol 26 September (Epub ahead of print). [DOI] [PubMed] [Google Scholar]
  56. Sohaib S., Cook G., Allen S., Hughes M., Eisen T., Gore M. (2009) Comparison of whole-body MRI and bone scintigraphy in the detection of bone metastases in renal cancer. Br J Radiol 82: 632–639 [DOI] [PubMed] [Google Scholar]
  57. Staudenherz A., Steiner B., Puig S., Kainberger F., Leitha T. (1999) Is there a diagnostic role for bone scanning of patients with a high pretest probability for metastatic renal cell carcinoma? Cancer 85(1): 153–155 [PubMed] [Google Scholar]
  58. Sun M., Ngo L., Genega E., Atkins M., Finn M., Rofsky N., et al. (2009) Renal cell carcinoma: dynamic contrast-enhanced MR imaging for differentiation of tumor subtypes correlation with pathologic findings. Radiology 250: 793–802 [DOI] [PubMed] [Google Scholar]
  59. Teo M., McDermott R.; Department of Medical Oncology, Adelaide and Meath Hospital, incorporating the National Children’s Hospital, Dublin, Ireland (2012) Does RECIST-defined progression correlate with lack of further sunitinib (SU) benefit in advanced renal cell carcinoma (aRCC)? J Clin Oncol 30(Suppl.): abstract e15093. [Google Scholar]
  60. Therasse P., Arbuck S., Eisenhauer E., Wanders J., Kaplan R., Rubinstein L., et al. (2000) New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 92: 205–216 [DOI] [PubMed] [Google Scholar]
  61. Ueno D., Yao M., Tateishi U., Minamimoto R., Makiyama K., Hayashi N., et al. (2012) Early assessment by FDG-PET/CT of patients with advanced renal cell carcinoma treated with tyrosine kinase inhibitors is predictive of disease course. BMC Cancer 12: 162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. van der Veldt A., Meijerink M., van den Eertwegh A., Haanen J., Boven E. (2010) Choi response criteria for early prediction of clinical outcome in patients with metastatic renal cell cancer treated with sunitinib. Br J Cancer 102: 803–809 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Williams R., Hudson J., Lloyd B., Sureshkumar A., Lueck G., Milot L., et al. (2011) Dynamic microbubble contrast-enhanced US to measure tumor response to targeted therapy: a proposed clinical protocol with results from renal cell carcinoma patients receiving antiangiogenic therapy. Radiology 260: 581–590 [DOI] [PubMed] [Google Scholar]
  64. Yuasa T., Urakami S., Yamamoto S., Yonese J., Nakano K., Kodaira M., et al. (2011) Tumor size is a potential predictor of response to tyrosine kinase inhibitors in renal cell cancer. Urology 77: 831–835 [DOI] [PubMed] [Google Scholar]
  65. Zisman A., Pantuck A., Dorey F., Said J., Shvarts O., Quintana D., et al. (2001) Improved prognostication of renal cell carcinoma using an integrated staging system. J Clin Oncol 19: 1649–1657 [DOI] [PubMed] [Google Scholar]

Articles from Therapeutic Advances in Medical Oncology are provided here courtesy of SAGE Publications

RESOURCES