Assessing the Response to Targeted Therapies in Renal Cell Carcinoma: Technical Insights and Practical Considerations

EUROPEAN UROLOGY 65 (2014) 766–777

available at www.sciencedirect.com journal homepage: www.europeanurology.com

Review – Renal Disease

Assessing the Response to Targeted Therapies in Renal Cell Carcinoma: Technical Insights and Practical Considerations Axel Bex a,*, Laure Fournier b, Nathalie Lassau c, Peter Mulders d, Paul Nathan e, Wim J.G. Oyen f, Thomas Powles g a

Department of Urology, The Netherlands Cancer Institute, Amsterdam, The Netherlands;

b

Universite´ Paris Descartes Sorbonne Paris Cite´, INSERM UMR-

S970, and Assistance Publique-Hoˆpitaux de Paris, Hoˆpital Europe´en Georges Pompidou, Service de Radiologie, Paris, France; c IRCIV, Institut Gustave Roussy, Villejuif, and IR4 M, UMR8081 Universite´ Paris-Sud 11, CNRS, Villejuif, France; d Department of Urology, Radboud University Medical Centre, Nijmegen, The Netherlands; e Mount Vernon Cancer Centre, Northwood, Middlesex, UK; f Department of Nuclear Medicine, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; g Queen Mary University of London, Barts and London School of Medicine, West Smithfield, London, UK

Article info

Abstract

Article history: Accepted November 18, 2013 Published online ahead of print on November 27, 2013

Context: The introduction of targeted agents for the treatment of renal cell carcinoma (RCC) has resulted in new challenges for assessing response to therapy, and conventional response criteria using computed tomography (CT) are limited. It is widely recognised that targeted therapies may lead to significant necrosis without significant reduction in tumour size. In addition, the vascular effects of antiangiogenic therapy may occur long before there is any reduction in tumour size. Objective: To perform a systematic review of conventional and novel imaging methods for the assessment of response to targeted agents in RCC and to discuss their use from a clinical perspective. Evidence acquisition: Relevant databases covering the period January 2006 to April 2013 were searched for studies reporting on the use of anatomic and functional imaging techniques to predict response to targeted therapy in RCC. Inclusion criteria were randomised trials, nonrandomised controlled studies, retrospective case series, and cohort studies. Reviews, animal and preclinical studies, case reports, and commentaries were excluded. A narrative synthesis of the evidence is presented. Evidence synthesis: A total of 331 abstracts and 76 full-text articles were assessed; 34 studies met the inclusion criteria. Current methods of response assessment in RCC include anatomic methods—based on various criteria including Choi, size and attenuation CT, and morphology, attenuation, size, and structure—and functional techniques including dynamic contrast-enhanced (DCE) CT, DCE-magnetic resonance imaging, DCEultrasonography, positron emission tomography, and approaches utilising radiolabelled monoclonal antibodies. Conclusions: Functional imaging techniques are promising surrogate biomarkers of response in RCC and may be more appropriate than anatomic CT-based methods. By enabling quantification of tumour vascularisation, functional techniques can directly and rapidly detect the biologic effects of antiangiogenic therapies compared with the indirect detection of belated effects on tumour size by anatomic methods. However, larger prospective studies are needed to validate early results and standardise techniques. # 2013 European Association of Urology. Published by Elsevier B.V. All rights reserved.

Keywords: Anatomic imaging Computed tomography Functional imaging Imaging Magnetic resonance imaging Positron emission tomography Renal cell carcinoma Targeted therapy Treatment response Ultrasonography

* Corresponding author. Department of Urology, The Netherlands Cancer Institute, Antoni van Leewenhoek Hospital, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands. Tel. +31 20 512 2553; Fax: +31 20 512 2554. E-mail address: [email protected] (A. Bex).

0302-2838/$ – see back matter # 2013 European Association of Urology. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.eururo.2013.11.031

EUROPEAN UROLOGY 65 (2014) 766–777

1.

Introduction

In the past few years, targeted agents that disrupt angiogenesis have been introduced for the treatment of metastatic renal cell carcinoma (mRCC). Approved agents include receptor tyrosine kinase inhibitors (TKIs), anti– vascular endothelial growth factor (VEGF) antibodies, and mammalian target of rapamycin inhibitors. Collectively, these agents have allowed for a substantial improvement in the treatment of the disease in terms of survival [1]. In contrast to tumour types in which biomarkers are used routinely to predict response to treatment, predictive biomarkers including imaging criteria are currently lacking in RCC. However, accurate and objective assessment of response is critical to ensure optimal use of targeted agents. Because antiangiogenic agents often cause decreased tumour vascularity and necrosis, traditionally used criteria involving measurement of tumour shrinkage (anatomic changes) may be inaccurate for assessing response to targeted agents [2]. Functional imaging, which tracks early changes in tumour physiology, may provide a more appropriate technique to monitor response to these therapies. Presently, there are no guidelines or general recommendations on the most suitable methods of response assessment for targeted therapy. We performed a systematic review of conventional and novel imaging methods for the assessment of response to targeted agents in RCC and discuss their use from a clinical perspective. 2.

Evidence acquisition

2.1.

Search strategy

2.3.

Data analysis

Baseline characteristics of studies included were collected for authors, types of studies, number of participants, types of imaging modalities, and outcome measures. A metaanalysis and an assessment of risk of bias were not planned due to the lack of randomised studies from a prior scoping exercise. A narrative synthesis of the evidence is presented instead. The Oxford Centre for Evidence Based Medicine 2011 level of evidence (OLoE) was used as a basis for the evidence synthesis (http://www.cebm.net/index.aspx?o= 5653). 3.

Evidence synthesis

The study selection process is outlined in the PRISMA diagram (Fig. 1). Thirty-seven studies met the inclusion criteria (3 phase 2 randomised studies, 19 nonrandomised comparative studies, 11 retrospective comparative studies, and 4 retrospective noncomparative studies). The baseline characteristics and OLoE of the included studies are displayed in Supplemental Table 1. 3.1.

Anatomic-based current methods of response assessment

3.1.1.

Computed tomography

Response to treatment has traditionally been based on measurements of tumour size reduction (30%) using the Response Evaluation Criteria in Solid Tumors (RECIST) [4]. In clinical practice, CT is the main technique used to evaluate RECIST response (Table 1) [4–7]. However, it can [(Fig._1)TD$IG]also be used to assess lesion attenuation, and degree and

The systematic review was performed according to Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines [3]. The databases searched were Medline (PubMed) and Google Scholar, covering the period from January 2006 to April 2013. Relevant articles were also identified using the related citations function of PubMed. In addition, abstracts from recent American Society of Clinical Oncology annual meetings and Genitourinary Cancers Symposia in 2011 and 2012 were searched. Additional sources of the search included the authors’ personal knowledge of the literature. The search terms included these terms: renal cell carcinoma, targeted therapy, imaging, ultrasound scanning, scintigraphy, magnetic resonance imaging (MRI), computed tomography (CT), and response and assessment (see Supplement for full Medical Subject Headings search). Only English-language articles were included. All abstracts and full-text articles were screened independently. Disagreement was resolved by discussion. 2.2.

Types of included study designs

Included were randomised controlled trials, nonrandomised controlled studies, retrospective case series, and cohort studies. Exclusion criteria were studies published before January 2006, systematic and narrative reviews, animal and preclinical studies, case reports, and commentaries.

767

Fig. 1 – Preferred Reporting Items for Systematic Reviews and Meta-analysis diagram outlining the study selection process.

768

EUROPEAN UROLOGY 65 (2014) 766–777

Table 1 – Summary of the available criteria for response assessment with computed tomography imaging of tumours Tumour response criteria

RECIST v.1.1 [4]

Method

Change in sum of longest dimensions of target lesions (maximum of two per organ; five total)

Tumour response assessment category CR

Disappearance of all target and nontarget lesions

PR

30% decrease in the sum of diameters of target lesions from baseline, and no progression of nontarget lesions or appearance of new lesions Insufficient change of target lesions to qualify for PR or PD, and no progression of nontarget lesions or appearance of new lesions 20% increase in the sum of diameters of target lesions over the smallest sum observed and overall 5-mm net increase, or unequivocal progression of nontarget lesions, or appearance of new lesions Disappearance of all lesions

SD

PD

Choi criteria [5]

Change in size or attenuation of target lesions

CR PR

SD PD

SACT criteria [6]

Change in size or attenuation of target lesions

Favourable response

Indeterminate response Unfavourable response

MASS criteria [7]

Change in size or attenuation of target lesions, together with assessment of tumour morphology and structure

Criteria for response

Favourable response

Indeterminate response Unfavourable response

A decrease in size* 10% or a decrease in tumour density (HU) 15% on CT, no new lesions, no obvious progression of nonmeasurable disease CR, PR, or PD criteria all unmet; no symptomatic deterioration attributed to tumour progression 10% increase in tumour size, tumour attenuation criteria for PR not met, new lesions, new intratumoural nodules or increase in the size of the existing intratumoural nodules No new lesions and one of the following: Decrease in tumour size* 20% Decrease in tumour size* 10% and 50% of nonlung target lesions with 20 HU decreased mean attenuation One or more nonlung target lesion with 40 HU decreased mean attenuation Insufficient change to qualify for favourable or unfavourable response Any of the following: Increase in tumour size* 20% New metastases, marked central fill-in** of a target lesion, or new enhancement in a homogeneously hypoattenuating nonenhancing mass No new lesions and one of the following: Decrease in tumour size* 20% One or more predominantly solid enhancing lesion with marked central necrosis or marked decrease attenuation (40 HUy)

Practical considerations

Limited assessment of morphology of lesions

Need for CECT

Need for CECT and 3D volumetric assessment, time consuming, labour intensive, subjective criteria for unfavourable response

Need for CECT (3D volumetric assessment not needed) Complex parameters for indeterminate or unfavourable response may limit use in routine clinical practice

Insufficient change to qualify for favourable or unfavourable response Any of the following: Increase in tumour size* 20% in the absence of marked central necrosis or marked decreased attenuation New metastases, marked central fill-in,z or new enhancement of a previously homogeneously hypoattenuating nonenhancing mass

3D = three dimensional; CECT = contrast-enhanced CT; CR = complete response; CT = computed tomography; MASS = morphology, attenuation, size, and structure; PD = progressive disease; PR = partial response; RECIST = Response Evaluation Criteria in Solid Tumors; SACT = size and attenuation CT; SD = stable disease. * Sum of longest diameters of target lesions as defined in RECIST v.1.1. ** Marked central fill-in defined as change from marked central necrosis to near-complete enhancement of solid tumour centrally. y Marked central necrosis is defined as >50% of the enhancing central portion of a predominantly solid enhancing mass subjectively changing to near-fluid attenuation (necrosis) after treatment. Marked decreased attenuation is defined as decreased attenuation of all or almost all of a mass by 40 HU using region-of-interest measurements of lesions on axial contrast-enhanced CT images. z Marked central fill-in is defined as a subjective change from marked central necrosis to complete or near-complete central intratumoural enhancement on contrast-enhanced CT.

EUROPEAN UROLOGY 65 (2014) 766–777

pattern of contrast enhancement. Response assessment to targeted agents in RCC is complicated because such therapies may lead to increased necrosis without tumour shrinkage [2]. Lesions can occasionally appear larger due to the development of necrosis, particularly in the liver. In addition, vascular effects of antiangiogenic therapy with changes in tumour density may precede, by a remarkably long interval, the effect on tumour size [8]. Many targeted agents result in clinical benefit but low RECIST response rates [9,10], although overall survival (OS) was significantly improved [1]. Therefore, response criteria are needed that better reflect changes associated with targeted therapy, such as tumour vascularity and necrosis, and that correlate with outcome. A retrospective analysis in 334 mRCC patients showed that a >10% decrease in the sum of the longest diameter (SLD) of target lesions on CT identified patients with a median progression-free survival (PFS) of 11.1 mo compared with 5.6 mo in those without a decrease following sunitinib treatment [11] (OLoE 4). These results were confirmed by two studies, one involving 70 mRCC patients treated with first-line sunitinib, sorafenib, or bevacizumab [12] (OLoE 4), and another with 196 patients receiving everolimus showing that thresholds from 0% to 10% predicted better PFS [13] (OLoE 4). The limitation of the use of this SLD threshold is that for small lesions, a 10% change in SLD may lie within the measurement error. To better assess necrosis as an indirect feature of reduced tumour vasculature, other CT-based criteria have been developed, such as the Choi criteria, which consider response as changes in tumour density or reduction of the longest tumour diameter [5]. These criteria were established for patients with gastrointestinal stromal tumours treated with imatinib [5], but they have also been applied to mRCC patients treated with sunitinib (n = 55) [14] (OLoE 4). Although in this study Choi criteria were better predictors of response than RECIST at first evaluation, their predictive value was similar to RECIST at later time points [14]. The authors concluded that although Choi criteria may help define at an early stage who is likely to benefit from sunitinib, their use will not change the management of these patients. In a study of 22 mRCC patients treated with sorafenib, Choi criteria defined more partial responses at early stages of therapy than RECIST. However, changes failed to predict prolonged OS [15] (OLoE 4) despite another study of 35 patients suggesting a correlation [8] (OLoE 4). Nathan et al. [16] (OLoE 4) modified the Choi criteria and used both changes in size and attenuation rather than either measure alone. They retrospectively evaluated CT scans from 20 mRCC patients treated with sunitinib or cediranib. Criteria combining a reduction in both size and arterial phase density correlated best with time to progression (TTP) compared with RECIST and Choi criteria. Other morphologic tumour response assessments that have been developed and evaluated include the size and attenuation computed tomography (SACT) criteria [6] (OLoE 4). However, in 84 mRCC patients treated with sorafenib or sunitinib, a module including morphology,

769

attenuation, size, and structure (MASS) proved more accurate in predicting TTP than RECIST, modified Choi, or SACT criteria [7] (OLoE 4). In addition, in a recent study, a combination of pretherapy clinical risk factors and CT response by MASS criteria effectively predicted PFS in mRCC patients treated with VEGF-targeted therapy [17] (OLoE 4). However, SACT and MASS methods rely on complex combinations of criteria, whereas response assessment should be simple to be suitable for routine use. Although the addition of tumour attenuation change to size criteria may improve response assessment, such analysis does not necessarily best reflect changes in tumour physiology. CT texture analysis, reflecting tumour heterogeneity, was shown to be independently associated with TTP [18] (OLoE 4). CT entropy decreased and uniformity increased after two cycles of TKI therapy. Further studies are required to ascertain whether the addition of this CT-based analysis to standard response assessment improves the prediction of response to TKIs in mRCC patients. 3.2.

Functional-based current methods of response assessment

3.2.1.

Dynamic contrast-enhanced computed tomography, magnetic

resonance imaging, and ultrasound

Functional techniques such as dynamic contrast-enhanced (DCE) imaging allow quantification of tumour vascularisation and can therefore directly detect the biologic effect of antiangiogenic therapies, rather than their belated indirect effect on tumour size [19]. DCE imaging follows the biodistribution of an intravenously injected contrast agent, transported by blood vessels to the tumour microcirculation and, according to its size, diffusing through the endothelial barrier into the tissue. Imaging follows the tracer distribution by quantifying signal enhancement in vessels and tissues over time [20]. Acquisitions of a series of images centred on a chosen lesion are repeated over time. Regions of interest (ROIs) placed on the tumour and a reference vessel (arterial input function) yield signal curves that can be modelled mathematically. Parameters are extracted that can be quantitative (eg, blood flow) or semiquantitative (eg, area under the curve [AUC]). Mean or median values can be calculated in the ROI, or a pixel-bypixel calculation of parameters can be depicted as parametric maps. These techniques can be applied using CT, MRI, or ultrasound (US). 3.2.2.

Dynamic contrast-enhanced computed tomography

DCE-CT has the advantage that it can be performed during regular follow-up examination of patients. It is more readily available and cheaper than MRI. Quantification is more straightforward than with MRI. An international working group published guidelines on how to perform DCE-CT to assess tumour vascularisation [19]. Disadvantages are the risk of iodine contrast nephropathy and the supplemental radiation dose. A perfusion acquisition adds approximately 10–20% of the radiation dose to a standard CT (ie, approximately 20 mSv) that must be taken into account when considering longitudinal follow-up.

770

EUROPEAN UROLOGY 65 (2014) 766–777

In mRCC, DCE-CT is able to predict and detect early response to targeted therapies. In a prospective study in 51 mRCC patients [20] (OLoE 2) treated with sorafenib versus placebo or sunitinib versus interferon a, DCE-CT yielded lower pretreatment parameters for future responders than for nonresponders (prediction of response). After a single therapy cycle, tumour blood flow and blood volume decreased by approximately 50% in patients receiving TKI but not in those on placebo or interferon (early detection of response). There was no decrease in size observed in the same time frame. These changes were not predictive of survival, however; the small number of patients did not allow sufficient statistical power. Large multicentric studies are currently ongoing to confirm these results. Figure 2 shows a DCE-CT with early response to targeted therapies before any decrease in tumour size. 3.2.3.

Dynamic contrast-enhanced magnetic resonance imaging

Studies have been performed to evaluate mRCC therapy using MRI, either with contrast agent injection (DCE) or [(Fig._2)TD$IG]

arterial spin labelling, where water protons in blood are magnetically tagged to serve as endogenous tracers [21–26]. MRI is nonradiating, and DCE sequences can be combined with an evaluation of the mobility of water molecules (diffusion-weighted MRI), MR spectroscopy evaluating molecular composition of tissues, or imaging of hypoxia (or blood oxygen level–dependent MRI). MRI constitutes a comprehensive one-stop shop, with a multiparametric evaluation of tumours and the effects of therapy. Conversely, disadvantages of MRI are that the implementation of standardised acquisitions is more challenging than for CT or US, and the conversion between signal intensity and concentration makes absolute quantification more difficult than CT [26] (OLoE 4). The most frequently used parameter in DCE-MRI, Ktrans, simultaneously reflects the blood flow (perfusion) and its passage into the extracellular space through the vascular endothelium (permeability). In a randomised trial of sorafenib versus placebo, Ktrans was a pharmacodynamic biomarker for sorafenib [24] (OLoE 2). However, variability was of concern. Changes after 4 wk of

Fig. 2 – Early response to sorafenib detected by dynamic contrast-enhanced computed tomography (CT). Transverse plane contrast-enhanced CT images of an adrenal metastasis in a patient with metastatic renal cell carcinoma (a) before therapy and (c) after 6 wk of therapy. On the corresponding parametric maps quantifying tissue blood volume, red pixels represent areas of high vascularity; blue areas represent poor vascularity. Before therapy, the size of the metastasis was 21 mm and the sum of the longest diameters of target lesions according to Response Evaluation Criteria in Solid Tumours (RECIST) was 115 mm. (b) Blood volume is measured at 10.7%. (d) On follow-up 6 wk later, high blood volume values (red pixels) have disappeared in the metastasis, with a mean blood volume measured at 3.3%, showing an early response according to functional imaging. The lesion, however, did not change size, measuring 18 mm, with a sum of the longest diameters of target lesions at 108 mm. The patient was therefore stable according to RECIST.

EUROPEAN UROLOGY 65 (2014) 766–777

sorafenib were not predictive of PFS, despite a smaller study suggesting a correlation [25] (OLoE 3). Results of the small studies are preliminary and suggest potential clinical application in the future [22–26]. 3.2.4.

Dynamic contrast-enhanced ultrasound

DCE-US can be performed using two different approaches that produce distinct results: bolus injection with time intensity curve (TIC) analysis and intravenous infusion with disruption replenishment analysis [27]. For the bolus injection analysis, single-plane imaging is usually performed at four frames per second over the duration of the ultrasound contrast agent (UCA) enhancement. The average intensity within the ROI can be displayed as a function over

771

time (ie, a TIC that describes the wash-in and wash-out of the contrast agent in the ROI). Most clinical studies are based on this method. The intravenous infusion with disruption replenishment analysis uses a pump or drip over 5–20 min. UCA is first imaged without disruption at a low mechanical index (MI); next, the MI is increased for a few frames, causing bubble disruption. Immediately following, the MI is returned to the nondisrupting level to observe the replenishment of the microbubbles into the ROI. This method was used in a study of RCC patients treated with sunitinib [27] (OLoE 3). Initially in mRCC, monitoring tumour response to targeted agents with contrast agents relied on qualitative analyses [28] (OLoE 2), but new methodologies using raw

[(Fig._3)TD$IG]

Fig. 3 – Early response to sunitinib assessed using dynamic contrast-enhanced ultrasonography (DCE-US) in a patient with metastatic renal cell carcinoma and an abdominal lesion. (a) B-mode, DCE-US, and corresponding computed tomography (CT) performed before treatment show high vascularity within the tumour. (b) DCE-US performed at day 7 demonstrates the beginning of necrosis within the tumour. (c) Contrast uptake curves with a strong decrease in area under the curve with a CT scan performed at 3 mo [31]. Reproduced with kind permission from Springer Science+Business Media: Current Urology Reports, Evaluation of treatment response in patients with metastatic renal cell carcinoma: role of state-of-the-art cross-sectional imaging, Volume 13, 2012, pp. 70–81, Katabathina VS, et al., Figure 3. Copyright Springer Science+Business Media, LLC 2011.

772

[(_)TD$FIG]

EUROPEAN UROLOGY 65 (2014) 766–777

Fig. 3. (Continued ).

linear data have been developed to produce robust and semiquantitative indexes. Analyses that include wash-in and wash-out times can be performed with curve fitting to determine functional parameters. The main parameters include peak intensity, AUC, area under the wash-in, area under the wash-out (all corresponding to blood volume), time to peak intensity (TPI), slope of the wash-in (SWI) (both corresponding to blood flow), and mean transit time. This technique enables imaging of the microcirculation, but no information on permeability can be obtained because of the pure blood pool nature of microbubbles [27]. Several studies using semiquantitative techniques with bolus injection have been published involving mRCC patients treated with sunitinib (n = 38) [29] (OLoE 3) but also other tumour types such as advanced hepatocellular carcinoma treated with bevacizumab (n = 42) [30]. In these studies, two parameters representing blood volume correlated with RECIST response. In the RCC study, TPI and SWI correlated with PFS and TPI with OS [29]. A patient with a response to sunitinib therapy using DCE-US is shown in Figure 3 [31].

In a French multicentre study including 19 centres with a total of 65 radiologists and 539 patients with various tumour types (n = 157 with mRCC), patients were treated with antiangiogenic therapies (sunitinib, sorafenib, and bevacizumab) [32,33] (OLoE 3). This multicentre study defined criteria that have an impact on the quality of DCE-US (quality 0–5). Only 3% of examinations with quality 0 were excluded from the final analysis [33]. Emerging evidence from >700 patients indicates that DCE-US may, with appropriate tools, differentiate between responders and nonresponders at an earlier stage than conventional methods [34]. This may allow tailoring of therapy, such as changing treatment in nonresponders or adapting dose according to toxicity [35]. DCE-US is recommended in European guidelines on the use of contrast-enhanced US to assess response to biologic therapy in mRCC, in dedicated centres with appropriate software for contrast signal quantification [36], and also in international guidelines [37]. However, DCE-US has several drawbacks. It is limited to lesions that are reproducibly detectable and is not a whole-body technique. DCE-US may therefore fail to

EUROPEAN UROLOGY 65 (2014) 766–777

[(Fig._4)TD$IG]

773

Fig. 4 – Response to sunitinib assessed using sequential [18F]-2-fluoro-2-deoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) in a patient with metastatic renal cell carcinoma. (A) Maximum intensity projection images of sequential whole-body PET scans. Baseline scan shows widespread FDG-avid metastatic lesions including multiple muscle deposits (arrowheads). The 4-wk scan shows a response in maximum standardised uptake value. The 16-wk therapy scan shows new FDG-avid muscle deposits consistent with progression. (B) Comparison of axial-fused PET-CT images at identical level through the chest at baseline and 4 wk after commencement of therapy showing a response to therapy. Adapted from Clinical Cancer Research 2011;17:6021–8, Kayani I, et al., Sequential FDG-PET/CT as a biomarker of response to sunitinib in metastatic clear cell renal cancer, with permission from AACR.

detect mixed responses or the appearance of new lesions, which may restrict its general use. 3.3.

Positron emission tomography

3.3.1.

(18F)-2-fluoro-2-deoxyglucose positron emission tomography

Studies suggest that early assessment of RCC using fluorine 18 (18F) fludeoxyglucose (FDG) positron emission tomography (PET)/CT (FDG-PET/CT) is predictive of response to targeted agents. However, not all RCC lesions accumulate 18 F-FDG, which limits the use of PET/CT for monitoring treatment response to those with FDG-avid lesions [38]. A preliminary study involving 12 mRCC patients treated with sunitinib demonstrated that early FDG-PET/CT findings after one cycle of sunitinib were consistent with later CT results [39] (OLoE 3). In a prospective phase 2 study of FDG-PET/CT in 44 mRCC patients treated with sunitinib, a metabolic response (ie, >20% reduction in maximum standardised uptake values [SUVmax]) occurred in 57% of patients after 4 wk, although this did not correlate with outcome [40] (OLoE 3) (Fig. 4). However, disease progression on FDG-PET/CT at week 16 was predictive of inferior survival. FDG-PET/CT at week 4 may have been too early in mRCC to predict resistance. Another prospective Japanese study involving 67 mRCC patients further supports that survival is predicted by SUVmax [41] (OLoE 3). Two separate studies have shown that early assessment by FDG-PET/CT of response to TKIs using a combination of FDG uptake and tumour size predicted both PFS and OS in mRCC patients (n = 30 and 12, respectively) [42,43] (OLoE 3). In addition, results from a prospective Norwegian study in 14 mRCC patients treated with sunitinib suggested that a high baseline 18F-FDG uptake may also indicate disease aggressiveness [44] (OLoE 3).

In primary RCC, FDG PET/CT of patients undergoing neoadjuvant therapy with sorafenib (n = 26) showed that primary clear cell RCC (ccRCC) tumours with lower SUV at baseline were more likely to respond to therapy [45] (OLoE 3). However, this trend was not observed for nonccRCC subtypes. 3.4.

Emerging methods of response assessment

3.4.1.

Other tracers for positron emission tomography

Although 18F-FDG is by far the most commonly used radiotracer in clinical practice, PET tracers depicting cellular features other than metabolism have been investigated for monitoring therapy response in mRCC. Assessment of tumour hypoxia using 18F-fluoromisonidazole PET/CT in 47 mRCC patients treated with sunitinib showed that PFS, but not OS, was shorter in those with hypoxic versus nonhypoxic metastases at baseline [46] (OLoE 3). In addition, 30 -deoxy-30 -(18F) fluorothymidine PET/CT has been used to characterise and quantify changes in tumour cell proliferation during sunitinib exposure and temporary withdrawal in patients with RCC and other solid malignancies [47] (OLoE 3). Further studies are required to determine the predictive value of these tracers. 3.4.2.

Radiolabelled monoclonal antibodies and peptides

Radiolabelled monoclonal antibodies and peptides developed for imaging in RCC have been investigated for diagnosis and staging or restaging [48,49]. The radiolabelled chimeric mAb iodine 124 (124I) girentuximab (also known as cG250), which specifically targets the cell-surface antigen carbonic anhydrase IX (CAIX) on ccRCC, has been studied most extensively [49,50] (OLoE 3). Clinical validation of 124I girentuximab immuno-PET demonstrated a

774

EUROPEAN UROLOGY 65 (2014) 766–777

Table 2 – A clinician’s viewpoint of the practical considerations of currently available imaging methods for response assessment to targeted agents in renal cell carcinoma CT

DCE-CT

DCE-MRI

DCE-US

PET

 Cost  No radiation  Simple standardised quantification using bolus injection  Largest clinical data set of functional imaging with strong evidence  No renal failure contraindication because of elimination by lung  Up to six injections feasible if necessary at each examination  Limited availability  Acquisition window restricted to 10 cm  15 cm during wash-in and wash-out  Not a whole-body technique  Bone, lung, and brain not evaluable

Imaging of metabolic activity

Advantages*

 Availability  Straightforward quantification  Long-term experience

 Availability  Straightforward quantification  Predictive of early response

 No radiation  Can be done without contrast agent injection (ASL)  Multiparametric evaluation (DCE, DWI)  Predictive of early response  Potentially predictive of OS

Disadvantages*

 Radiation dose  Limited in patients with renal insufficiency  Not predictive for early response

 Radiation dose 10–20% higher than standard CT  Limited in patients with renal insufficiency  Predictive value for OS not yet established

    

No standardised acquisition Quantification more complex Cost Availability Predictive value for OS not yet validated in larger studies  Limited experience

Availability Cost Radiation dose Low sensitivity Very limited experience  Data not validated     

ASL = arterial spin labelling; CT = computed tomography; DCE = dynamic contrast-enhanced; DWI = diffusion-weighted imaging; MRI = magnetic resonance imaging; OS = overall survival; PET = positron emission tomography; US = ultrasound. * Methodology, ease of use, availability of equipment, experience to date, and costs were considered.

sensitivity and specificity of 86.2% and 85.9% for ccRCC, the subtype most responsive to targeted agents [50]. Although preliminary evidence suggests that CAIX expression may be a predictive biomarker for response to sorafenib [51], the potential role of 124I girentuximab imaging in predicting response to targeted drugs in ccRCC remains to be established. The potential of indium 111 (111In)-labelled bevacizumab scintigraphy to image VEGF-A expressing ccRCC tumours was also investigated in nine patients who received neoadjuvant sorafenib before nephrectomy [48] (OLoE 3). Neoadjuvant sorafenib resulted in a significant decrease of 111In bevacizumab uptake in RCC. Nevertheless, VEGF could still be demonstrated in the tumours, indicating that sorafenib prevented penetration of bevacizumab into the tumours. Similar observations have been made when studying the effects of sorafenib on 111In girentuximab targeting of ccRCC [52]. These results suggest a potential role for radiolabelled agents to design rational dosing regimens when considering combinations of targeted drugs, but further research is required. 3.5.

Selecting appropriate imaging techniques for response

assessment to targeted therapies

As RCC treated with targeted agents undergoes both morphologic and functional changes, it is likely that in the future imaging involving a combination of parameters or methods including CT, MRI, DCE-US, and PET may be used to monitor therapy. The ultimate promise of functional and

molecular imaging is to predict response very early after initiation of therapy or, better still, prior to commencement of targeted therapy. Such imaging biomarkers could then be used to tailor treatment to patients who are most likely to benefit from therapy, thereby preventing significant toxicities and the administration of ineffective treatment, and reducing costs for the health care system. More immediately, scientifically and analytically validated imaging biomarkers have the potential to assist in drug development if incorporated into rationally designed hypothesis-testing clinical trials. For example, imaging biomarkers could be used to confirm that drugs are reaching their target, to evaluate progressive disease, or to serve as intermediate decision end points for phase 2 trials. Therefore, in both clinical trials and routine practice, assessment of early response to targeted therapy will ultimately be more important than detecting RECIST progression, which may only show after significant latency. Response assessment needs to be performed easily in a standardised fashion, across multiple institutions and countries. Table 2 summarises the relative merits/limitations of the currently used anatomic and functional imaging methods. To date, definite validation of functional and molecular imaging techniques as established surrogate end points of response is lacking. Multiple retrospective and smaller prospective studies support low-level evidence that anatomic- and functional-based imaging techniques are associated with predicting early response to targeted therapy and, in some instances, outcome.

EUROPEAN UROLOGY 65 (2014) 766–777

Given the large number of mRCC patients evaluated, the strongest evidence exists for DCE-US in predicting early therapy responders. However, DCE-US does not allow whole-body exploration and therefore requires a complementary technique to explore every potential metastatic site. For CT, the 10% threshold for response has shown to be predictive of survival in large series; however, it is not clear how this might change patient management. More important than the choice of specific anatomic- and functional-based imaging techniques is the lack of randomised phase 3 evaluation of these modalities in patients receiving antiangiogenic therapy. We do not currently know if absence of early response to therapy on imaging requires early switch of treatment and if such early treatment changes impact on outcome, quality-of-life parameters, and health care costs. Without this evidence to guide us in treatment decisions, no clinical recommendation can be made for the use of a specific imaging method to detect early response. Based on currently available modalities, it is difficult to imagine a predictive imaging biomarker used before treatment that would accurately reflect disease biology sufficiently to inform treatment choice in the clinic, or one with sufficient widespread availability and expertise in interpretation to allow routine monitoring of response at an early stage during treatment. Therefore, tumour size remains, by default, the most important criteria for routine clinical use. Based on current evidence and experience, CT will likely remain the most common modality of choice for the foreseeable future. This is because CT is whole body, relatively inexpensive, and widely available with standardised protocols. 4.

Conclusions

Accurate and practical methods of response assessment are critical for the optimal use of targeted therapies in clinical practice. Ensuring the accuracy of response assessment will also have important consequences for the design of future clinical trials. Available evidence shows that functional imaging techniques are promising surrogate biomarkers of response in RCC and may be more appropriate than anatomic-based methods such as size-based CT assessment. However, larger prospective studies are needed to validate early results and standardise the imaging techniques before they can be used to help determine individualised treatment strategies in RCC.

775

Statistical analysis: Bex, Fournier, Lassau. Obtaining funding: Bex. Administrative, technical, or material support: Bex, Fournier, Lassau. Supervision: Bex, Fournier, Lassau, Mulders, Nathan, Oyen, Powles. Other: None. Financial disclosures: Axel Bex certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Axel Bex has participated in advisory boards for Pfizer, Bayer, GSK, and Novartis. He is the principal investigator of the EORTC SURTIME trial, supported in part by an educational grant from Pfizer. Nathalie Lassau has received a grant from the Institut National de Cancer for the DCE-US STIC study, and received payment from Pfizer, Novartis, Hoffmann-La Roche, Bracco, and Toshiba for lectures including service on a speakers’ bureau. Peter Mulders has no conflicts of interest to disclose. Paul Nathan has received payment for advisory boards and speaker fees from GSK, Pfizer, BMS, Novartis, Roche, and Bayer. Laure Fournier has received honoraria from Pfizer. Wim J.G. Oyen is a consultant for Wilex AG, Munich, Germany. Thomas Powles has received educational grants from Pfizer and GSK, and he has participated on advisory boards for Pfizer, Astellas, and GSK. Funding/Support and role of the sponsor: Pfizer Inc. funded the study and was involved in review and approval of the manuscript. Acknowledgment statement: Medical writing support was provided by Tracey Spurway and Simon Hampson (both freelance medical writers) and Rachel Mason at ACUMED (Tytherington, UK).

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.eururo.2013.11.031. References [1] Hutson TE. Targeted therapies for the treatment of metastatic renal cell carcinoma: clinical evidence. Oncologist 2011;16(Suppl 2): 14–22. [2] Van der Veldt AA, Meijerink MR, van den Eertwegh AJ, Boven E. Targeted therapies in renal cell cancer: recent developments in imaging. Target Oncol 2010;5:95–112. [3] Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. [4] Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228–47.

Author contributions: Axel Bex had full access to all the data in the study

[5] Choi H, Charnsangavej C, Faria SC, et al. Correlation of computed

and takes responsibility for the integrity of the data and the accuracy of

tomography and positive emission tomography in patients with

the data analysis.

metastatic gastrointestinal stromal tumor treated at a single institution with imatinib mesylate: proposal of new computed tomog-

Study concept and design: Bex, Fournier, Lassau, Mulders, Nathan, Oyen, Powles.

raphy response criteria. J Clin Oncol 2007;25:1753–9. [6] Smith AD, Lieber ML, Shah SN. Assessing tumor response and

Acquisition of data: Bex, Fournier, Lassau, Mulders, Nathan, Oyen, Powles.

detecting recurrence in metastatic renal cell carcinoma on targeted

Analysis and interpretation of data: Bex, Fournier, Lassau, Mulders,

therapy: importance of size and attenuation on contrast-enhanced

Nathan, Oyen, Powles. Drafting of the manuscript: Bex, Fournier, Lassau, Oyen.

CT. AJR Am J Roentgenol 2010;194:157–65. [7] Smith AD, Shah SN, Rini BI, Lieber ML, Remer EM. Morphology,

Critical revision of the manuscript for important intellectual content: Bex,

attenuation, size, and structure (MASS) criteria: assessing response

Fournier, Lassau, Mulders, Nathan, Oyen, Powles.

and predicting clinical outcome in metastatic renal cell carcinoma

776

EUROPEAN UROLOGY 65 (2014) 766–777

on antiangiogenic targeted therapy. AJR Am J Roentgenol 2010;194: 1470–8. [8] Schmidt N, Hess V, Zumbrunn T, Rothermundt C, Bongartz G, Potthast S. Choi response criteria for prediction of survival in patients with metastatic renal cell carcinoma treated with antiangiogenic therapies. Eur Radiol 2013;23:632–9. [9] Escudier B, Eisen T, Stadler WM, et al. Sorafenib in advanced clearcell renal-cell carcinoma. N Engl J Med 2007;356:125–34. [10] Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med 2007;356:115–24.

[24] Hahn OM, Yang C, Medved M, et al. Dynamic contrast-enhanced magnetic resonance imaging pharmacodynamic biomarker study of sorafenib in metastatic renal carcinoma. J Clin Oncol 2008; 26:4572–8. [25] Flaherty KT, Rosen MA, Heitjan DF, et al. Pilot study of DCE-MRI to predict progression-free survival with sorafenib therapy in renal cell carcinoma. Cancer Biol Ther 2008;7:496–501. [26] Kang HC, Tan KS, Keefe SM, et al. MRI assessment of early tumor response in metastatic renal cell carcinoma patients treated with sorafenib. AJR Am J Roentgenol 2013;200:120–6.

[11] Thiam R, Fournier LS, Trinquart L, et al. Optimizing the size varia-

[27] Williams R, Hudson JM, Lloyd BA, et al. Dynamic micro-bubble

tion threshold for the CT evaluation of response in metastatic renal

contrast-enhanced US to measure tumor response to targeted

cell carcinoma treated with sunitinib. Ann Oncol 2010;21:936–41.

therapy: a proposed clinical protocol with results from renal cell

[12] Krajewski KM, Guo M, Van den Abbeele AD, et al. Comparison of

carcinoma patients receiving antiangiogenic therapy. Radiology

four early posttherapy imaging changes (EPTIC; RECIST 1.0, tumor

2011;260:581–90.

shrinkage, computed tomography tumor density, Choi criteria) in

[28] Lamuraglia M, Escudier B, Chami L, et al. To predict progression-free

assessing outcome to vascular endothelial growth factor-targeted

survival and overall survival in metastatic renal cancer treated with

therapy in patients with advanced renal cell carcinoma. Eur Urol

sorafenib: pilot study using dynamic contrast-enhanced Doppler

2011;59:856–62.

ultrasound. Eur J Cancer 2006;42:2472–9.

[13] Oudard S, Thiam R, Fournier LS, et al. Optimisation of the tumour

[29] Lassau N, Koscielny S, Albiges L, et al. Metastatic renal cell carcino-

response threshold in patients treated with everolimus for meta-

ma treated with sunitinib: early evaluation of treatment response

static renal cell carcinoma: analysis of response and progression-

using dynamic contrast-enhanced ultrasonography. Clin Cancer

free survival in the RECORD-1 study. Eur J Cancer 2012;48:1512–8.

Res 2010;16:1216–25.

[14] Van der Veldt AA, Meijerink MR, van den Eertwegh AJ, Haanen JB,

[30] Lassau N, Koscielny S, Chami L, et al. Advanced hepatocellular

Boven E. Choi response criteria for early prediction of clinical

carcinoma: early evaluation of response to bevacizumab therapy

outcome in patients with metastatic renal cell cancer treated with

at dynamic contrast-enhanced US with quantification—preliminary

sunitinib. Br J Cancer 2010;102:803–9.

results. Radiology 2011;258:291–300.

[15] Hittinger M, Staehler M, Schramm N, et al. Course of size and density

[31] Katabathina VS, Lassau N, Pedrosa I, Ng CS, Prasad SR. Evaluation of

of metastatic renal cell carcinoma lesions in the early follow-up of

treatment response in patients with metastatic renal cell carcino-

molecular targeted therapy. Urol Oncol 2012;30:695–703.

ma: role of state-of-the-art cross-sectional imaging. Curr Urol Rep

[16] Nathan PD, Vinayan A, Stott D, Juttla J, Goh V. CT response assess-

2012;13:70–81.

ment combining reduction in both size and arterial phase density

[32] Lassau N, Chami L, Chebil M, et al. Dynamic contrast-enhanced

correlates with time to progression in metastatic renal cancer

ultrasonography (DCE-US) and anti-angiogenic treatments. Discov

patients treated with targeted therapies. Cancer Biol Ther 2010; 9:15–9.

Med 2011;11:18–24. [33] Lassau N, Chapotot L, Benatsou B, et al. Standardization of dy-

[17] Smith AD, Shah SN, Rini BI, Lieber ML, Remer EM. Utilizing pre-

namic contrast-enhanced ultrasound for the evaluation of anti-

therapy clinical schema and initial CT changes to predict progres-

angiogenic therapies. The French Multicenter Support for

sion-free survival in patients with metastatic renal cell carcinoma

Innovative and Expensive Techniques Study. Invest Radiol

on VEGF-targeted therapy: a preliminary analysis. Urol Oncol 2013; 31:1283–91. [18] Goh V, Ganeshan B, Nathan P, Juttla JK, Vinayan A, Miles KA. Assessment of response to tyrosine kinase inhibitors in metastatic renal cell cancer: CT texture as a predictive biomarker. Radiology 2011;261:165–71. [19] Miles KA, Lee TY, Goh V, et al., Experimental Cancer Medicine

2012;47:711–6. [34] Lassau N, Cosgrove D, Armand JP. Early evaluation of targeted drugs using dynamic contrast-enhanced ultrasonography for personalized medicine. Future Oncol 2012;8:1–4. [35] Michels J, Lassau N, Gross-Goupil M, Massard C, Mejean A, Escudier B. Sunitinib inducing tumor lysis syndrome in a patient treated for renal carcinoma. Invest New Drugs 2010;28:690–3.

Centre Imaging Network Group. Current status and guidelines

[36] Piscaglia F, Nolsøe C, Dietrich CF, et al. The EFSUMB Guidelines and

for the assessment of tumour vascular support with dynamic

Recommendations on the Clinical Practice of Contrast Enhanced

contrast-enhanced computed tomography. Eur Radiol 2012;22:

Ultrasound (CEUS): update 2011 on non-hepatic applications.

1430–41.

Ultraschall Med 2012;33:33–59.

[20] Fournier LS, Oudard S, Thiam R, et al. Metastatic renal carcinoma:

[37] Claudon M, Dietrich CF, Choi BI, et al. Guidelines and good clinical

evaluation of antiangiogenic therapy with dynamic contrast-

practice recommendations for contrast enhanced ultrasound

enhanced CT. Radiology 2010;256:511–8.

(CEUS) in the liver—update 2012: a WFUMB-EFSUMB initiative

[21] Notohamiprodjo M, Sourbron S, Staehler M, et al. Measuring perfusion and permeability in renal cell carcinoma with dynamic contrast-enhanced MRI: a pilot study. J Magn Reson Imaging 2010; 31:490–501. [22] Desar IM, ter Voert EG, Hambrock T, et al. Functional MRI techni-

in cooperation with representatives of AFSUMB, AIUM, ASUM, FLAUS and ICUS. Ultrasound Med Biol 2013;39:187–210. [38] Dilhuydy MS, Durieux A, Pariente A, et al. PET scans for decisionmaking in metastatic renal cell carcinoma: a single-institution evaluation. Oncology 2006;70:339–44.

ques demonstrate early vascular changes in renal cell cancer

[39] Vercellino L, Bousquet G, Baillet G, et al. 18F-FDG PET/CT imaging

patients treated with sunitinib: a pilot study. Cancer Imaging 2012;

for an early assessment of response to sunitinib in metastatic renal

11:259–65.

carcinoma: preliminary study. Cancer Biother Radiopharm 2009;

[23] de Bazelaire C, Alsop DC, George D, et al. Magnetic resonance

24:137–44.

imaging-measured blood flow change after antiangiogenic therapy

[40] Kayani I, Avril N, Bomanji J, et al. Sequential FDG-PET/CT as a

with PTK787/ZK 222584 correlates with clinical outcome in meta-

biomarker of response to sunitinib in metastatic clear cell renal

static renal cell carcinoma. Clin Cancer Res 2008;14:5548–54.

cancer. Clin Cancer Res 2011;17:6021–8.

777

EUROPEAN UROLOGY 65 (2014) 766–777

[41] Namura K, Minamimoto R, Yao M, et al. Impact of maximum stan-

[47] Liu G, Jeraj R, Vanderhoek M, et al. Pharmacodynamic study using

dardized uptake value (SUVmax) evaluated by 18-fluoro-2-deoxy-D-

FLT PET/CT in patients with renal cell cancer and other solid

glucose positron emission tomography/computed tomography (18F-

malignancies treated with sunitinib malate. Clin Cancer Res 2011;

FDG-PET/CT) on survival for patients with advanced renal cell carcinoma: a preliminary report. BMC Cancer 2010;10:667.

17:7634–44. [48] Desar IM, Stillebroer AB, Oosterwijk E, et al.

111

In-bevacizumab

[42] Ueno D, Yao M, Tateishi U, et al. Early assessment by FDG-PET/CT of

imaging of renal cell cancer and evaluation of neoadjuvant treat-

patients with advanced renal cell carcinoma treated with tyrosine

ment with the vascular endothelial growth factor receptor inhibitor

kinase inhibitors is predictive of disease course. BMC Cancer 2012; 12:162.

sorafenib. J Nucl Med 2010;51:1707–15. [49] Divgi CR, Pandit-Taskar N, Jungbluth AA, et al. Preoperative char-

[43] Minamimoto R, Nakaigawa N, Tateishi U, et al. Evaluation of

acterisation of clear-cell renal carcinoma using iodine-124-labelled

response to multikinase inhibitor in metastatic renal cell carcinoma

antibody chimeric G250 (124I-cG250) and PET in patients with

by FDG PET/contrast-enhanced CT. Clin Nucl Med 2010;35:918–23.

renal masses: a phase I trial. Lancet Oncol 2007;8:304–10.

[44] Revheim ME, Winge-Main AK, Hagen G, Fjeld JG, Fossa˚ SD, Lilleby

[50] Divgi CR, Uzzo RG, Gatsonis C, et al. Positron emission tomography/

W. Combined positron emission tomography/computed tomogra-

computed tomography identification of clear cell renal cell

phy in sunitinib therapy assessment of patients with metastatic

carcinoma: results from the REDECT trial. J Clin Oncol 2013;31:

renal cell carcinoma. Clin Oncol (R Coll Radiol) 2011;23:339–43.

187–94.

[45] Khandani AH, Cowey CL, Moore DT, Gohil H, Rathmell WK. Primary

[51] Choueiri TK, Regan MM, Rosenberg JE, et al. Carbonic anhydrase IX

renal cell carcinoma: relationship between 18F-FDG uptake and

and pathological features as predictors of outcome in patients with

response to neoadjuvant sorafenib. Nucl Med Commun 2012;33:

metastatic clear-cell renal cell carcinoma receiving vascular endo-

967–73.

thelial growth factor-targeted therapy. BJU Int 2010;106:772–8.

[46] Hugonnet F, Fournier L, Medioni J, et al. Metastatic renal cell carci-

[52] Muselaers CHJ, Stillebroer AB, Boers-Sonderen MJ, et al. Sorafenib

noma: relationship between initial metastasis hypoxia, change after

reduces the tumor uptake of Indium-111-girentuximab in clear cell

1 month’s sunitinib, and therapeutic response: an 18F-fluoromiso-

renal cell carcinoma patients. Eur J Nucl Med Mol Imaging 2012;

nidazole PET/CT study. J Nucl Med 2011;52:1048–55.

39(Suppl 2):S188–9.