Usefulness of magnetic resonance imaging in prostate cancer

Radiología. 2010;52(6):513-524 ISSN: 0033-8338

RADIOLOGÍA

RADIOLOGÍA

Publicación Oficial de la Sociedad Española de Radiología Médica Incluida en Index Medicus/MEDLINE

www.elsevier.es/rx

Actividad acreditada en base a la encomienda de gestión concedida por los Ministerios de Educación, Cultura y Deporte y de Sanidad y Consumo al Consejo General de Colegios O ciales de Médicos con 1 crédito, equivalente a 4 horas lectivas. www.seram.es

www.elsevier.es/rx

UPDATE

sefulness of magnetic resonance imaging in prostate cancer J.C. Vilanovaa b c

J. Cometd R. arcia Figueirase J. Barcelóa b c M. Boadaa

Unidad de Resonancia Magn tica, l nica irona, irona, Spain Servicio de Radiodiagnóstico, Hospital Sta aterina, Salt, irona, Spain c epartamento de iencias M dicas, acultad de Medicina, Universidad de irona, irona, Spain d Servicio de Urolog a, Hospital Universitario r rueta, irona, Spain e Servicio de Radiodiagnóstico, Hospital l nico Universitario, Santiago de ompostela, Spain a

b

Received 25 November 2009; accepted 4 June 2010

KEYWORDS Prostate tumors; Magnetic resonance imaging; Spectroscopy; Diffusion MRI; Contrast agent.

Abstract In the last decade, technical advances in magnetic resonance imaging (MRI) have made it the technique of choice in the overall management of patients with suspected or conrmed prostate cancer. MR makes it possible to acquire information about morphology and function in the same examination by using techniques like spectroscopy, diffusion, and dynamic sequences with intravenous contrast material administration. Moreover, MRI enables both focused study of the prostate gland and of regional and/or whole-body involvement, depending on the clinical indications, in less than an hour. The main clinical indications for MRI of the prostate are a) staging local, regional, and/or remote disease; b) detecting prostate cancer or guiding prostate biopsy in cases of clinical suspicion or negative ndings in previous biopsy specimens; and c) monitoring the response to treatment. It is important to know the different protocols with speci c MRI sequences for the prostate, depending on the different clinical indications, to ensure that they are performed and interpreted correctly. This article provides up-to-date information about the use of MRI for the study of the prostate to show how the morphological and functional information can be used in clinical practice. © 2009 SERAM. Published by Elsevier España, S.L. All rights reserved.

PALABRAS CLAVE Neoplasia de próstata; Imagen por resonancia magnética; Espectroscopia; IRM difusión;

tilidad de la resonancia magn tica en el cáncer de próstata Resumen Los avances técnicos de la resonancia magnética (RM) en la última década hacen que se considere la técnica de elección en el manejo global del paciente con sospecha o diagnóstico de cáncer de próstata. La RM permite combinar información morfológica y funcional al mismo

Corresponding author. E-mail: [email protected] (J.C. Vilanova). 0033-8338/$ - see front matter © 2009 SERAM. Published by Elsevier España, S.L. All rights reserved.

514

Medio de contraste

J.C. Vilanova et al tiempo mediante la aplicación de secuencias como la espectroscopia, difusión y secuencias dinámicas con contraste endovenoso en el mismo estudio. La RM permite no solo focalizar el estudio en la glándula sino valorar también la extensión regional a toda la pelvis o a todo el cuerpo dependiendo de la indicación clínica, en menos de 1 hora de exploración. Las principales indicaciones clínicas de la RM de próstata son: a) estadi cación local, regional o a distancia; b) detección o guía para biopsia diagnóstica ante la sospecha clínica o resultado negativo en biopsias previas; y c) monitorización terapéutica. Es preciso conocer los distintos protocolos con secuencias especí cas en RM de próstata, dependiendo de las diferentes indicaciones clínicas, para su correcta realización e interpretación. Este artículo pretende ser una actualización en la utilización de la RM de próstata en el manejo del cáncer de próstata, describiendo un enfoque útil que permita aplicar la información morfológica y funcional en la práctica clínica. © 2009 SERAM. Publicado por Elsevier España, S.L. Todos los derechos reservados.

Introduction Prostate cancer is a large-scale public health problem and the most common malignancy in men in industrialised countries. Despite these data, there are no accurate tools currently available for prostate cancer diagnosis 1. Clinical suspicion of prostate cancer is based on digital rectal examination (DRE) and on elevated prostate-speci c antigen (PSA). There are two forms of PSA in the serum: free form and attached to proteins. PSA produced by normal tissue is less likely to bind to proteins, while neoplastic tissue produces more of the attached form. Thus, patients with a free-to-total PSA ratio > 30 % are rather unlikely to have cancer, whereas for ratio < 25 % the probability increases 2. There is no consensus regarding the pathological level of total PSA, although values higher than 4 ng/ml are considered suspicious. The accepted approach with PSA > 4 ng/ml is to take a prostate biopsy3. Excessive use of PSA levels entails a great number of unnecessary biopsies; with 60–70 % of biopsies with negative results4. Low positive predictive value of PSA is due to overlap with benign prostatic pathology and prostatitis, where elevated PSAmay also be found. In addition, the presence of multiple negative biopsies before establishing a de nite diagnosis of cancer is not infrequent5. Traditionally, imaging techniques have been of little relevance for prostate cancer management. Only transrectal ultrasound (US) as guide for biopsy, and computerized tomography (CT) for abdominopelvic staging have been 6 used, but with limited accuracy for local or regional staging . Incidentally, the new technological advances in MR have proved useful to detect cancer in patients with clinical suspicion (change in PSA values, free PSA ratio or rate of PSA increase) and to attain a more reliable local, regional or systemic staging 7. Technological progress of MR provides a detailed morphological image in high resolution that is used as a map and guidance to direct biopsy using transrectal US, increasing significantly the diagnostic accuracy in the detection and localization of cancer 8 . In addition to anatomical information, MR imaging provides metabolic

information by means of spectroscopy (MR spectroscopy), molecular information, using diffusion MR imaging (DMR), and also information about the vascularization using dynamic sequences obtained after the intravenous administration of a contrast agent (CMR) 9. The possibility to integrate this information allows not only locating the lesion, but also determining the degree of differentiation or aggressiveness of the tumor 10. Today, it is possible to perform a complete MR study for the different clinical indications by applying adequately different protocols. Although prostate MR is not yet widely used, it is being used routinely in certain institutions and clinical environments. In this paper, we review and update the role of RM in the management of patients with clinical suspicion or diagnosis of prostate cancer and we describe a practical approach for implementing and integrating morphological and functional information into clinical practice.

MR e amination technique Equipment and preparation MR examination of the prostate needs high eld units with no lower than 1.5T. Nonetheless, 3T units — recently introduced — offer a better signal-to-noise ratio 11. An optimal MR study for localization and detection of prostate cancer requires the use of an endorectal coil and a pelvic multi-channel phased-array coil; endorectal coil is more uncomfortable and makes the exploration more expensive. Depending on the clinical indications, e.g. staging of a previously diagnosed prostate cancer, a morphological study with a pelvic multi-channel phased-array coil may be enough. For 3T units, the need of using an endorectal coil has not been determined yet 11,12 . Preparation for the endorectal study requires bland diet 12h before the study and avoiding stimulating drinks such as caffeine and theine. Enema or glucagon administration may be used, but it is not indispensable. Bladder repletion is not needed; in fact, given the duration of the examination, it is better if the

Usefulness of magnetic resonance imaging in prostate cancer

patient urinates before to avoid having to urinate during the procedure. A DRE using urological lubricant is done before the introduction of the US MR probe in order to prepare the sphincter and rule out rectal stenosis which may hamper anal probing. Next, the probe is in ated with 80–100 cc of

515

liquid air. Different liquids have been used such as barium sulphate suspension that reduces susceptibility artifacts11. If there is a history of previous biopsies, MR examination has to be postponed 8–10 weeks to avoid that glandular hemorrhagic changes (fig. 1) or periglandular fibrotic changes may interfere with the correct interpretation of the study13,14.

Morphological MR imaging. Anatomy Morphological sequences consist of spin-echo (SE) or fast spin-echo (FSE) acquisitions (table 1).

Figure 1 Anatomy in the axial plane, T1-weighted sequence. A) Diffuse normal isointense signal of the gland, without differentiation between central and peripheral zone, and some high intensity areas of adenoma of no pathological signi cance (arrow). B) High intensity signal in both peripheral lobes secondary to postbiopsy hemorrhage (arrows) with asymmetry of the neurovascular bundle of non tumoral origin.

1. Transverse T1-weighted FSE images were obtained from the aortic bifurcation to the symphysis pubis using the following parameters: repetition time (RT), 400–500 ms; echo time (ET), minimum; section thickness, 5 mm; intersection gap, 1 mm < field of view (FOV), 34 cm; matrix, 256 × 192; number of excitations, 1; and frequency direction, transverse to prevent motion artifact from the intestinal loops. This sequence allows assessment of pelvic lymph nodes and of the pelvis bone to rule out metastasis and to assess hemorrhagic changes after biopsy (table 2) ( g. 1). 2. T2-weighted FSE images in the three planes were obtained from the prostate and seminal vesicles. RT, 6000–7500 ms; ET, 96–130 ms; echo train length, 16; section thickness, 3 mm; intersection gap, 0 mm; FOV, 14–16 cm; matrix, 256 × 192, frequency direction, antero-posterior (in axial acquisition); and number of excitations, three. This sequence allows the evaluation of the normal prostate anatomy (table 2). Interpretation has not been proved better using fat saturation imaging15. The prostate consists of four zones: peripheral posterior or peripheral gland, central, transitional and anterior bro stromal. The central and transitional zones form

Table 1 Acquisition parameters of the prostate MR protocol in our 1.5T unit Sequence

T1 (SE)

T2 (FRFSE)

T2 (FRFSE)

DWI

GE FAST SPGR

PRESS

Plane Coil Anatomical coverage ET (ms) RT (ms) Section thickness (mm) FOV (cm) Matrix Acquisitions Echo train Acquisition time Flip angle ( ) Bandwith b-value (s/mm 2)

Axial ATD-TORSO Aortic bifurcation -pubic symphysis Minimum 400–500 5

Axial ATD-TORSO Prostate-seminal vesicles 130 7000–7500 3

Coronal/sagittal ATD-TORSO Prostate-seminal vesicles 96 6000–6500 3

Axial ATD-TORSO Iliac crest- pubic symphysis Minimum 6000–6500 5

Axial 3D ATD-TORSO Prostate-seminal vesicles 2 14 4

Axial ENDO-ATD Prostate 130 1000 29.8

34 256 × 192 1 — 3:26 — 20.83 —

14 256 × 192 3 16 4:21 — 20.83 —

16 256 × 192 3 16 4:01 — — —

36 128 × 80 6 — 2:32 — — 0.1000

26 256 × 160 1 — 3:35 12 — —

12 16 × 8 1 — 18:52 — — —

ATD-TORSO: integrated (ATD) endorectal (ENDO) + pelvic coil (TORSO) connection; DWI: diffusion weighted imaging; FRFSE: fast recovery fast spin echo; FSE: fast spin echo; GE FAST SPGR: echo gradient fast spoiled gradient; SE: spin echo; PRESS: point resolved spectroscopy.

516

J.C. Vilanova et al

Table 2 Advantages and disadvantages of the different MR sequences Imaging type

Advantages

Disadvantages

MR

Allows anatomical and functional images of the prostate MR is more accurate in lesion detection and staging than any other method

Expensive

T1 MR

Detection of hemorrhages post-biopsy as zones of high intensity. Detection of nodes and bone lesions

Prostate gland appears homogeneous

T2 MR

Differentiation of anatomical zones

Prostatitis, hemorrhages, atrophy, BPH and post-treatment changes may look like cancer Central gland tumors signal is similar to the normal gland signal and to the central gland hypertrophy Transitional zone hyperplasia with cystic and brotic nodes may show similar signal pattern as cancer Lymph nodes staging is complicated and should be based on their morphology and enlargement

Does not allow real-time imaging

Cancer shows low signal intensity Peripheral zone tumors appear as not well de ned foci of low intensity Allows assessment of ECE SVI appears as foci of low signal intensity within the high signal intensity of the seminal vesicles

CMR

Detection of vascularization Increased speci city in comparison with T2 MR only; early tumoral enhancement, with rapid washout of contrast agent

Small low grade tumors may not show increased vascularization Patients with BPH may show an abnormal increase of the pattern similar to that of patients with central gland tumors Contrast agent with gadolinium may caused nephrogenic systemic brosis in patients with severe renal failure

MRS

Provides concentrations of citrate, choline and creatine; high levels of choline and low levels of citrate may be indicative of cancer

Technological challenge and may required physicists expert in MR. Overlap between the metabolic pro le of prostatitis and cancer

DMR

Prostate cancer may appear as a focus of high intensity and then as a low signal in the ADC map

Findings may not be speci c since hyperplasia may also have low diffusion. Susceptibility problems due to postbiopsy hemorrhage

ADC: apparent diffusion coef cient; ECE: extracapsular extension; BPH: benign prostatic hyperplasia; SVI: seminal vesicles invasion; CMR: dynamic contrast-enhanced MR imaging; DMR: diffusion-weighted MR; MRS: MR spectroscopy; T1 MR: T1-weighted MR; T2 MR: T2-weighted MR.

the central gland 16. In adulthood, benign prostatic hypertrophy arises from the transitional zone, compressing the central zone and forming the pseudo-capsule or surgical capsule (fig. 2). The transitional zone is called either central zone or central gland in adulthood. The peripheral gland shows homogeneous high signal intensity in contrast to the heterogeneous low signal intensity of the central-transitional zone, although high intensity areas of adenoma may be observed in the central zone. The

neurovascular bundle is located postero-laterally to the peripheral gland-zone ( g. 2). The diagnostic criterion for prostate cancer on T2-weighted MR images is the presence of nodular areas of low signal intensity within the normal high signal intensity of the peripheral gland or zone ( g. 3). T2-weighted imaging has limitations for identifying cancer in the transitional zone due to the dif culty to de ne tumoral zones of low signal intensity within the normal low signal intensity of the central tissue. The ndings suggesting cancer in the transitional

Usefulness of magnetic resonance imaging in prostate cancer

Figure 2 Axial schematic shows the anatomy of the prostate in an adult with correlation with axial T2-weighted images. High signal intensity of the normal peripheral zone (PZ). Central zone (CZ) or surgical capsule–pseudocapsule (white arrows). Transitional zone (TZ) with normal heterogeneous hypertrophy of low signal intensity. True capsule (white arrows heads). Neurovascular bundle (black arrows heads). Fibro stromal anterior zone (AZ).

Figure 3 Axial T2–weighted and spectroscopy image. Normal spectrum shows elevation of citrate and low level of choline with high signal intensity of the peripheral zone (thin arrow). Pathological spectrum with elevated choline and low citrate from a low signal intensity lesion of the left peripheral zone related to carcinoma (thick arrow).

Figure 4 T2–weighted imaging of non neoplastic lesions of low signal intensity in the peripheral zone. A) Bilateral diffuse peripheral low intensity caused by prostatitis (arrows). B) Bilateral diffuse peripheral low intensity caused by subacute hemorrhage after biopsy (arrows). C) Focal lesion of low signal intensity in the right peripheral zone caused by brosis (arrow). D) Diffuse low signal intensity of the entire prostate secondary to brachytherapy. Implantation seeds (arrows).

517

zone are: diffuse low signal intensity with poorly de ned margins, absence of a margin with (commonly seen in adenomas) and disruption of the surgical pseudocapsule8. Low signal intensity lesions in the peripheral zone may also be observed in benign lesions such as prostatitis, hyperplasia, brosis, subacute post-biopsy hemorrhage, post-radiation therapy changes, and hormone deprivation therapy (fig. 4). The prostate does not have a real capsule, but a fibro-muscular band named prostatic capsule ( g. 2). More than 70 % of prostate cancers arise in the peripheral gland and the rest in the central gland. In prostates with large hypertrophic areas in the transitional zone, where the peripheral tissue is compressed, cancer localization may be dif cult using systematic blind biopsies7.

MR spectroscopy MR spectroscopic imaging provides information on the gland metabolism 17. Sequence is acquired using the multivoxel technique, programming a volume on T2-weighted imaging and according to the guidance parameters speci ed for our MR unit (table 1). For three-dimensional spectroscopic acquisition we use selective bands to achieve water and lipid suppression and to optimize citrate and lipid choline detection. Six saturation-bands are used to eliminate signals from the adjacent tissue. Field homogenization is automatically calculated. Spatial resolution of 0.24– 0.34 cm 3, RT/ET: 1000/130 ms, 16 × 8 × 8 phase-encoded spectral arrays, and 19 min acquisition time. The dataset should be processed using speci c software for quanti cation of the values of the metabolic ratios. Generally, the peaks calculated are for choline, creatine and citrate. Despite other metabolites can be also identi ed such as polyamines, they are better detected in 3T units 18. Good signal-to-noise ratio needs to be con rmed (> 5:1) as well as the absence of contamination by fat or water. Normal prostatic tissue has high citrate values and low choline and creatine values; whereas neoplastic tissue has high choline values and low citrate values ( g. 3). Suspicion criterion is a ratio (creatine + choline)/citrate (CCo/Ci) > 0.7 obtained in the study of the peripheral gland in a 1.5T unit 19; however, there is no consensus regarding which metabolic ratio determines the presence of prostate cancer, due to the variability among patients and MR units9. Parameters of the metabolic ratio for the central gland have not been established due to the overlap in CCo/Ci ratio between the normal, or hypertrophic, and the neoplastic tissue in the central zone 20. In any case, the most speci c nding for diagnosis of neoplasia in the central zone is the absence or very low values of citrate and the elevation of choline 21.

Diffusion MR The technological advances of MR have allowed an expansion in the use of the diffusion sequence (DMR) to organs other than the brain, since spectroscopic imaging was rst used for diagnosis of stroke. Diffusion sequence provides information about random brownian motion of the free water molecules in the interstitial space and through the cellular membrane. In general, neoplastic

518

J.C. Vilanova et al

tissue has more restricted diffusion than normal tissue because of its higher cellular density, which hampers normal diffusion of water molecules22. Diffusion sequence provides information about cellular density, the tortuosity of the extracellular space, the integrity of the cellular membranes, and the degree of glandular organization. For our unit, the diffusion sequence is spin echo—single shot echo planar image (TR/TE: 6000–6500/minimum ms, b value: 0 and 1.000 s/mm 2 in the axial plane) (table 1). The sequence must include the whole pelvis, prostate and seminal vesicles in order to simultaneously achieve regional staging and detect a potential glandular lesion in the same acquisition. The low mobility of molecules appears as a high signal on DMR; similarly, molecules with high mobility show a loss of signal. Sequence interpretation requires processing and quantification of the diffusion images using the ADC (apparent diffusion coef cient) in the parametric map (fig. 5). For quantification, the 5–10 mm 2 region of interest (ROI) is placed on the region of choice. Factor b value varies since there is no consensus regarding its optimal value 23. Nonetheless, values = 0 and ≥ 1000 s/mm2 are recommended for the prostate. The use of higher b values may increase the sensitivity of the sequence by eliminating the high signal intensity of tissues with long T2 relaxation times (edema or uid due to the high density of their protons), phenomenon known as “T2 shine-through” Lesions with serious diffusion restrictions appear as low signals on grey-scale ADC maps or blue on color maps ( g. 5). There is no a clearly de ned ADC value to differentiate between cancer and absence of cancer; however, values < 1.2 × 10 -3 mm 2/s are a quite indicative threshold of neoplastic process 24. In order to avoid misinterpretations, diffusion images need to be evaluated along with anatomical images. Advantages of diffusion are short acquisition time and high contrast resolution between tumor and normal tissue. However, its

drawbacks are poor spatial resolution and the potential susceptibility artifacts caused by postbiopsy hemorrhage (table 2).

Figure 5 Cancer in the central zone of the prostate. Focal lesion of low signal intensity in the left central zone at axial T2-weighted imaging. MR spectroscopy sequence shows elevated choline peak (arrow). DMR shows slight unspeci c high signal intensity (arrow). ADC color map shows low ADC values at lesion level in blue color (arrow); ADC value may be quantify by placing the region of interest (ROI).

Figure 6 Prostate cancer in central and peripheral zones. CMR sequence shows the parametric map (arrow) with high enhancement during the rst seconds seen in yellow and rapid washout in the intensity-time curve seen in red. MR spectroscopy sequences show elevated choline (white arrow) and low ADC value that appears blue in the color parametric map (black arrow).

Dynamic MR Dynamic contrast-enhanced MR imaging (CMR) allows evaluation of tumor vascularization and also indirectly of the angiogenesis 25. For the prostate, a T1-weighted 3D gradient-echo sequence of the whole gland and seminal vesicles is normally used, with a high temporal resolution. In our unit, we use a 3D fast gradient-echo sequence (table 1) (TR/TE: 14/2; angle: 12 ; thickness: 4 mm; FOV: 26; matrix: 256 × 160) with a 4–9 s temporal resolution.The sequence is repeated until reaching a 3–5 min total length. Analysis of the values of the contrast signal intensity over time ratios can be done in three ways: qualitative (curve pro le), semi-quantitative (changes in the signal intensity) or quantitative 22. Qualitative values measure the types of curve pro le: type I (progressive enhancement), type II (in plateau) or type III (rapid washout) similar to perfusion imaging studies of breast. Semi-quantitative values measure and quantify the relative signal intensity (higher postcontrast signal intensity/prescontrast signal ratio), slope of the intensity-time curve (that shows speed of enhancement) or the area under the intensity-time curve. These measurements are easy to obtain in the workstations, but cannot be compared among different units ( g. 6). For obtaining quantitative parameters, pharmacokinetics parameters are used, which allow quanti cation of various

Usefulness of magnetic resonance imaging in prostate cancer

parameters: ktrans (transit of the contrast from the vascular compartment to the interstice through the endothelium), kep (return to the vascular compartment) and Ve (fraction of extracellular space of tumor) 26. In addition, using these data we can generate parametric maps representing the intratumoral heterogeneity of the vascular distribution. Nonetheless, we should take into account the complexity behind these parameters, the lack of standardization and of universal post-processing software. In areas where the tumor shows high vascular permeability (as the peripheral zone), ktrans values depend mainly on the ow; whereas in the centre of the tumor (where permeability is the limiting factor), ktrans values depend on the permeability surface 27. Interpretation difficulties arise from the lack of standardization. However, some authors have proved that the most reliable parameter for neoplasm detection is a high enhancement slope during the rst seconds along with a rapid washout rate14,28.

Whole body MR imaging Today, advances in MR technology allow performing a whole body study for the screening of bone metastases in less than 30 min. The examination includes: diffusion sequence in the axial plane of all the stations, whole-body coronal T1-weighted sequence and STIR; and sagittal T1-weighted sequence of the whole spine. Detailed description of the protocol and of the whole-body technique is beyond the scope of this review, but this information can be obtained from the specific referenced publication about the whole-body MR technique and diffusion sequence29.

Clinical indications Diagnosis and localization Guides do not recommend yet MR for prostate cancer detection 30. However, MR usefulness has proved a tool to guide the biopsy to a focal area in patients with elevated PSA and previous negative biopsies 31. MR has proved more useful for prostate cancer detection than DRE o systematic blind biopsy 32. MR is also useful for cancer detection in the peripheral zone as well as in the central zone ( g. 5), which is actually a dif cult access zone from where samples are not usually obtained in routine biopsies21. High resolution T2-weighted imaging is relatively sensitive but not very speci c for prostate cancer localization 33, as described in the examination technique section (fig. 4) (table 2). Combination of T2-weighted sequence with one or two functional sequences, MR spectroscopy, DMR or CMR (table 2), may improve cancer detection22,34-39. The optimal combination of functional sequences to complement T2-weighted images and the efficacy combining all functional sequences has not been proven. Integration of DMR imaging into the study protocol should be done routinely, since its usefulness has been demonstrated if combined with T2-weighted imaging39,40, without increasing the cost of the procedure, unlike spectroscopy or dynamic contrast-enhanced MR imaging. Integration of the morphological and functional information in a single MR study points to an improvement in the diagnostic capability

519

of the technique (fig. 6), particularly in the central transitional zone 28,41. In fact, MR diagnostic accuracy of cancer for the central gland is lower than that for the peripheral gland. This is due to the presence of similar ndings in morphological MR and functional MR in benign hypertrophy, prostatitis and cancer 41,42 . Functional technique has the advantage of providing information to predict the degree of tumor differentiation, which correlates with the values of the speci c tumor markers for prostate cancer 43,44. Preliminary studies seem to put in objective terms the correlation between elevated values of tumor markers and functional imaging parameters that are pathological and suggestive of more differentiated neoplasias. Recently, it has been demonstrated the usefulness of combining the integrated information in MR with clinical parameters such as free PSA levels or PSA density (PSA value-prostate size ratio) in order to select the patients who are good candidates for biopsy, in contrast with the current indication of undergoing biopsy for patients with PSA > 4 ng/ml 37,45. The idea is to select patients with high clinical risk of cancer in order to perform a MR examination, to perform more accurate biopsies in patients with tumor suspicion on functional MR.These new proposed algorithms would obviate a significant number of unnecessary biopsies in patients with elevated PSA, due to the high negative predictive value of MR for prostate cancer. Moreover, performing a prebiopsy MR would avoid artifacts caused by postbiopsy hemorrhage or brosis that may interfere in the correct interpretation of the prostate MR study (table 2)13,14. Likewise, the use of MR as prebiopsy tool allows obtaining information about tumor localization to help us direct the needle into the lesion. However, it is complicated to transfer the MR information to the US screen upon biopsy30. To that end there are different options. First, we can specify tumor localization by using a morphological diagram, although this approach does hold operator-related errors. Second, we can use a coregistration of the MR image on the US screen in real time; nonetheless, it is technically dif cult to integrate a static image, such as a MR image, into a dynamic US image 46 . Third, we can perform a MR-guided biopsy in the MR room; however, availability, costs and complexity of the procedure may be a problem due to the amount of equipment needed. Probably, the most feasible option, and there are already some prototypes, is to coregistrate the MR image on the US screen upon biopsy. One of the main dif culties in the diagnostic assessment of the prostate is to differentiate between chronic prostatitis and cancer. In both cases, the peripheral gland shows low signal intensity at T2-weighted sequences ( g. 7). In the MR spectroscopic sequence, the CCo/Ci ratio appears elevated, similar to what happens in neoplastic tissue 47. In dynamic contrast-enhanced sequences overlap may also occurs since hypervascularization appears in both prostatitis and cancer 28. Some findings suggest prostatitis, particularly bilateral diffuse or patchy low signal intensity of well defined margins or triangular at T2-weighted imaging ( g. 7). Diffusion sequence may be useful to differentiate prostatitis from prostate cancer. Chronic prostatitis generally has intermediate ADC values, unlike the lowerADC values of cancer; nonetheless, there is overlap inADC values

520

Figure 7 Chronic prostatitis in a 45 years old patient with 17 ng/ml PSA. Axial T2-weighted image shows bilateral patchy low signal intensity in the peripheral zone with regular margins. MR spectroscopy sequence shows elevated choline peaks with elevated (choline + creatine)/citrate (CCo/Ci) ratio. Parametric map does not show low ADC values and there are no focal areas in blue. CMR sequence shows bilateral hypervascularization in zones of suspicion at the axialT2 image, predominantly on the right 1. Histological examination showed chronic prostatitis in all the biopsy cylinders.

J.C. Vilanova et al

It is important to be aware that capsule irregularity may be observed after biopsy ( g. 10) 13, but this fact does not mean extension to the capsule or postbiopsy artifacts. For instance, a previous study reported a decrease in the accuracy of local staging from 83 % to 46 % due to MR postbiopsy artifact 52. It is dif cult to objectively assess the data published on MR reliability in prostate cancer local staging, due to the different techniques and methods employed. However, a meta-analysis performed on 74 studies on prostate cancer staging concluded that the best results are obtained with high-resolution MR multiplanar imaging using endorectal coil49. The most speci c criterion for extracapsular extension (ECE) is the asymmetry of the neurovascular bundle (38 % sensibility, 95 % specificity) and the most sensitive is the overall assessment of all the criteria of ECE (68% sensitivity, 72 % speci city) 53. The studies that have analyzed the selection criteria for MR studies on prostate cancer staging suggest including patients with an intermediate clinical risk ofT3 stage (PSA, 10–20 ng/ml; Gleason 5–7) 54. This approach has a good cost-effectiveness ratio for patients with intermediate risk and with high risk of ECE (PSA> 20 ng/ml and Gleason > 7).

between benign hypertrophy and cancer24 (table 2), specially in the central gland39.

Staging The TNM classi cation is the reference standard for prostate cancer staging. The aim is to determine the anatomical extent of cancer and to differentiate intraglandular from locally invasive or metastatic lesions, in order to decide treatment: curative (prostatectomy, brachytherapy, cryotherapy, HIFU High Intensity Focused Ultrasound ) or palliative (radiotherapy, hormonotherapy)48. Cancer without extraglandular extension is considered T2 stage and T3 when it spreads outside the capsule. Due to its high contrast and spatial resolution, MR is the most reliable method for prostate cancer staging, in comparison with computed tomography (CT), echography and DRE 49,50. MR criteria for extracapsular extension (ECE) of a prostatic tumor are 7,51 ( g. 8): irregular and spiculated focal bulging of the capsule, loss of the normal low signal intensity of the capsule, obliteration of the rectoprostatic angle, asymmetry and involvement of the neurovascular bundle; and tumor extension into the seminal vesicles (table 3).The invasion of the seminal vesicle is demonstrated by the presence of low signal within the vesicles 51. Diffusion sequence has proved useful as a parameter of seminal vesicle in ltration46 ( g. 9). In order to obtain the maximum accuracy of MR in prostate cancer staging, it is essential to use an endorectal or a multichannel pelvic coil to obtain high-resolution studies of the pelvis11.

Figure 8 MR criteria for ECE. A) Irregular and spiculated focal bulging of the capsule (arrow). B) Obliteration of the rectoprostatic angle (arrow). C) Involvement and asymmetry of the neurovascular bundle and periprostatic fat (arrow). D) Extension into the left seminal vesicle (arrow).

Table 3 MR criteria for ECE Irregular focal bulging and spiculated of the capsule Loss of the normal low signal intensity of the capsule Obliteration of the rectoprostatic angle Involvement or asymmetry of the neurovascular bundle, periprostatic fat Extension into the seminal vesicles

Usefulness of magnetic resonance imaging in prostate cancer

Figure 9 Diagnosis and staging using MR only. 63 years old patient with 6 ng/ml PSA and 7 % free PSA with no previous biopsies. A) Diffuse low signal intensity of the left peripheral lobe (arrow). B) Low signal intensity of the root of the seminal vesicle suggestive of in ltration (arrow), seen in blue on the parametric ADC map (arrow) indicative of restricted diffusion with values < 1.2 × 10 —3 mm 2/s (C) (b factor: 1000 s/mm 2). D) Slight low signal intensity of the right pubic branch and high signal intensity on the DMR meaning restricted diffusion (arrow) (E), related to metastasis.

521

agreement regarding which node size is considered pathological, but a maximum value of 8 mm for round nodes and 10 mm for oval ones, measured in the short axis, is accepted. In order to improve reliability, organ-specific contrast agents have been developed. Ultrasmall superparamagnetic particles of iron oxide (USPIO) are injected intravenously and taken up by macrophages of the reticuloendothelial system of the node, causing a decrease in the MR signal intensity. Metastatic nodes, where the macrophages are replaced by the tumor, do not capture the iron particles and, as a result, there is no modi cation of the MR signal. Initial experience suggests that USPIO used in MR lympography improves the sensitivity and speci city for detection of metastatic nodes 57. In any case, the use of this organ-speci c contrast has not been approved in the clinical practice. Bone metastases screening is recommended in patients with PSA > 20 ng/ml or with high levels of alkaline phosphatase in blood 58. Although the technique for bone metastases detection is bone scintigraphy , the development of the MR technology makes possible to perform a whole-body MR study during the study of local staging, using the same coil antenna without moving the patient. The addition of the diffusion sequence of the whole pelvis to the examination protocol of prostate cancer allows a more accurate diagnosis of bone lesions ( g. 9). Results show that MR imaging is more accurate than bone scintigraphy 59. Maybe in the future, a whole body MR will be performed as the method of choice instead of bone scintigraphy, when there will be more MR resources and availability. The option of screening for bone metastases using only a MR study of the spine is also considered 60.

Therapeutic monitoring

Figure 10 Postbiopsy changes. Bilateral prostate neoplasia (long arrow) with spiculated left capsular region (short arrow) four weeks after biopsy, that it may be suggestive of ECE. The histopathological exam revealed intracapsular neoplasia.

The added value of endorectal MR in intraglandular prostate cancer prediction has been demonstrated for all risk groups, but particularly for high risk groups, when compared with stage prediction using the Partin tables53,55, which are based exclusively on clinical and histological parameters. Despite the excellent contrast of soft tissues at MR imaging, the method has limited value in the assessment of nodal metastasis; with reliability similar to that of CT. In any case, combination of the Partin tables with the overall MR information has proved useful to assess ECE, invasion of the seminal vesicles and nodal metastasis due to the high predictive value for nodal metastasis56. There is no common

Nowadays, the post therapeutic monitoring method for diagnosis of tumoral relapse is elevated PSA, which is considered a biochemical relapse 58. MR imaging can be useful for detection and localization of local relapse f o r s u r g i c a l t r e a t m e n t 61, r a d i o b a c h y t h e r a p y 62, hormonotherapy63, cryotherapy64,65 or High-Intensity Focused Ultrasound (HIFU)66. After different treatments, MR ndings may be dif cult to interpret due to the presence of glandular atrophy and brosis caused by radiation, brachytherapy seeds, scar or post-surgical clips (table 2). T2-weighted sequence has low sensitivity for tumoral relapse detection due to a diffuse low signal intensity caused by atrophy 67 that prevents differentiating neoplastic tissue from atrophic tissue ( g. 4). In these cases, it is advisable to use functional sequences such as MR spectroscopy, DMR and CMR to improve the detection of the lesion61,62,68 ( g. 11). Conventional MR sequences have proved useful for localization of potential post-prostatectomy relapse, although an endorectal coil is required 69. T2-weighted sequence and CMR are effective for assessing the response to HIFU treatment ( g. 11)66. Diffusion sequence may differentiate tumoral relapse from post-radiation changes based on the different ADC values; however, its role in prostate cancer management is yet to be determined 23.

522

J.C. Vilanova et al

Authorship J.C. Vilanova has made contributions to the conception and design of the article. All the authors have been involved in drafting the paper or in its critical review, making important intellectual contributions. All the authors have approved the final version of the manuscript that is submitted for publication.

Acknowledgement We would like to thank Dr. M.D. Figueras for her invaluable collaboration.

Figure 11 Relapse after HIFU treatment in a 69 years old patient with 1.2 ng/ml PSA, one year after HIFU. A) Glandular atrophy with diffuse low signal intensity (arrow) atT2-weighted sequence. MR spectroscopy shows elevated choline peak (arrow) suggestive of neoplasia. DMR shows low value seen in blue on the parametric ADC map (arrow). Intensity-time curve shows an elevated slope due to early enhancement related to tumoral relapse.

Future prospects The clear ongoing progress in the development and application of MR technology in the management of prostate cancer suggests that it will be included in the routine clinical practice. Platforms of computer-aid design (CAD) are now also being developed and will allow integration of all the information gathered from the RM studies in one screen. In other words, we have at our disposal the morphological information at T2-weighted sequences, combined with functional information from the spectroscopy, diffusion and vascular studies ( g. 6). One technological challenge is to improve parameters standardization and interpretation of the different sequences obtained from the different units available in the market. In addition, we should have the tool to transfer the MR information to the US screen to perform the prostate biopsy ef ciently.

Conclusion Prostate MR imaging is the technique of choice for prostate cancer management in the diagnosis, staging and therapeutic monitoring. The use of MR as a prebiopsy technique in patients with risk of prostate cancer has advantages and a signi cant bene t in patient management. Functional MR imaging allows improving the accuracy of the method, and an understanding of its advantages and disadvantages is required for a correct interpretation (table 2). Ongoing progress and improvement of the technology should increase the efficacy for patient management so that MR could be included in clinical practice guidelines.

References 1. Borras J, Marcos-Gragera R, Torres A, Espinas JA. Analysis of cancer incidence, survival and mortality according to the main tumoral localizations, 1985–2019: prostate cancer . Med Clin (Barc). 2008;131(Suppl 1):63-6. 2. Roddam AW, Duffy MJ, Hamdy FC, Ward AM, Patnick J, Price CP, et al. Use of prostate-speci c antigen (PSA) isoforms for the detection of prostate cancer in men with a PSA level of 2–10 ng/ml: systematic review and meta-analysis. Eur Urol. 2005;48:386-99. 3. Andriole GL, Catalona WJ. Using PSA to screen for prostate cancer. The Washington University experience. Urol Clin North Am. 1993;20:647-51. 4. Shah JB, Reese AC, McKiernan JM, Benson MC. PSA updated: still relevant in the new millennium? Eur Urol. 2005;47:427-32. 5. Arcangeli CG, Ornstein DK, K eetch DW , Andriole GL. Prostate-speci c antigen as a screening test for prostate cancer . The United States experience. Urol Clin North Am. 1997;24: 299-306. 6. Engeler CE, Wasserman NF, Zhang G. Preoperative assessment of prostatic carcinoma by computerized tomography . Weaknesses and new perspectives. Urology. 1992;40:346-50. 7. Turkbey B, Albert PS, Kurdziel K, Choyke PL. Imaging localized prostate cancer: current approaches and new developments. Am J Roentgenol. 2009;192:1471-80. 8. Hricak H, Choyke PL, Eberhardt SC, Leibel SA, Scardino PT . Imaging prostate cancer: a multidisciplinary perspective. Radiology. 2007;243:28-53. 9. Choi YJ, Kim JK, Kim N, Kim KW, Choi EK, Cho KS. Functional MR imaging of prostate cancer. Radiographics. 2007;27:63-75. 10. Shukla-Dave A, Hricak H, Kattan MW, Pucar D, Kuroiwa K, Chen HN, et al. The utility of magnetic resonance imaging and spectroscopy for predicting insigni cant prostate cancer: an initial analysis. BJU Int. 2007;99:786-93. 11. Mazaheri Y, Shukla-Dave A, Muellner A, Hricak H. MR imaging of the prostate in clinical practice. MAGMA. 2008;21:379-92. 12. Beyersdorff D, Taymoorian K, K nosel T, Schnorr D, Felix R, Hamm B, et al. MRI of P rostate Cancer at 1.5 and 3.0T : Comparison of Image Quality in Tumor Detection and Staging. Am J Roentgenol. 2005;185:1214-20. 13. Qayyum A, Coakley FV, Lu Y, Olpin JD, W u L, Yeh BM, et al. Organ-con ned prostate cancer: effect of prior transrectal biopsy on endorectal MRI and MR spectroscopic imaging. Am J Roentgenol. 2004;183:1079-83. 14. Tamada T, Sone T, Jo Y, Yamamoto A, Yamashita T, Egashira N, et al. P rostate cancer: relationships between postbiopsy hemorrhage and tumor detectability at MR diagnosis. Radiology. 2008;248:531-9.

Usefulness of magnetic resonance imaging in prostate cancer 15. Tsuda K, Yu KK, Coakley FV, Srivastav SK, Scheidler JE, Hricak H. Detection of extracapsular extension of prostate cancer: role of fat suppression endorectal MRI. J ComputAssist Tomogr. 1999;23:74-8. 16. McNeal JE. The zonal anatomy of the prostate. Prostate. 1981; 2:35-49. 17. V ilanova JC, Barcelo J. P rostate cancer detection: magnetic resonance (MR) spectroscopic imaging. Abdom Imaging. 2007; 32:253-61. 18. K urhanewicz J, Vigneron DB. Advances in MR spectroscopy of the prostate. Magn Reson Imaging Clin N Am. 2008;16:697-x 19. K urhanewicz J, V igneron DB, Hricak H, Narayan P , Carroll P, Nelson SJ. Three-dimensional H-1 MR spectroscopic imaging of the in situ human prostate with high (0.24–0.7-cm 3) spatial resolution. Radiology. 1996;198:795-805. 20. Akin O, Sala E, Moskowitz CS, Kuroiwa K, Ishill NM, Pucar D, et al. Transition zone prostate cancers: features, detection, localization, and staging at endorectal MR imaging. Radiology. 2006;239:784-92. 21. Zakian KL, Eberhardt S, Hricak H, Shukla-Dave A, Kleinman S, Muruganandham M, et al. Transition zone prostate cancer: metabolic characteristics at 1H MR spectroscopic imaging–initial results. Radiology. 2003;229:241-7. 22. Somford DM, Futterer JJ, Hambrock T, Barentsz JO. Diffusion and perfusion MR imaging of the prostate. Magn Reson Imaging Clin N Am. 2008;16:685-95. ix 23. Padhani AR, Liu G, Koh DM, Chenevert TL, Thoeny HC, Takahara T, et al. Diffusion-weighted magnetic resonance imaging as a cancer biomarker: consensus and recommendations. Neoplasia. 2009;11:102-25. 24. Gibbs P, Tozer DJ, Liney GP , Turnbull LW. Comparison of quantitative T2 mapping and diffusion-weighted imaging in the normal and pathologic prostate. Magn R eson Med. 2001;46: 1054-8. 25. Alonzi R, Padhani AR, Allen C. Dynamic contrast enhanced MRI in prostate cancer. Eur J Radiol. 2007;63:335-50. 26. Ocak I, Bernardo M, Metzger G, Barrett T, Pinto P, Albert PS, et al. Dynamic contrast-enhanced MRI of prostate cancer at 3T: a study of pharmacokinetic parameters. Am J R oentgenol. 2007;189:849. 27. Van Dorsten FA, van der GM, Engelbrecht MR, van Leenders GJ, Verhofstad A, Rijpkema M, et al. Combined quantitative dynamic contrast-enhanced MR imaging and (1)H MR spectroscopic imaging of human prostate cancer. J Magn Reson Imaging. 2004;20:279-87. 28. Engelbrecht MR, Huisman HJ, Laheij RJ, Jager GJ, van Leenders GJ, Hulsbergen-Van De Kaa CA, et al. Discrimination of prostate cancer from normal peripheral zone and central gland tissue by using dynamic contrast-enhanced MR imaging. Radiology . 2003;229:248-54. 29. V ilanova JC, Barcelo J. Diffusion-weighted whole-body MR screening. Eur J Radiol. 2008;67:440-7. 30. Ahmed HU, Kirkham A, Arya M, Illing R, Freeman A, Allen C, et al. Is it time to consider a role for MRI before prostate biopsy? Nat Rev Clin Oncol. 2009;6:197-206. 31. Amsellem-Ouazana D, Younes P, Conquy S, Peyromaure M, Flam T, Debre B, et al. Negative prostatic biopsies in patients with a high risk of prostate cancer. Is the combination of endorectal MRI and magnetic resonance spectroscopy imaging (MRSI) a useful tool? A preliminary study. Eur Urol. 2005;47:582-6. 32. Comet-Batlle J, V ilanova-Busquets JC, Saladie-R oig JM, Gelabert-Mas A, Barcelo-Vidal C. The value of endorectal MRI in the early diagnosis of prostate cancer. Eur Urol. 2003;44: 201-7. 33. Vilanova JC, Comet J, Capdevila A, Barcelo J, Dolz JL, Huguet M, et al. The value of endorectal MR imaging to predict positive biopsies in clinically intermediate-risk prostate cancer patients. Eur Radiol. 2001;11:229-35.

523 34. Futterer JJ, Heijmink SW, Scheenen TW, Veltman J, Huisman HJ, Vos P, et al. P rostate cancer localization with dynamic contrast-enhanced MR imaging and proton MR spectroscopic imaging. Radiology. 2006;241:449-58. 35. R einsberg SA, P ayne GS, Riches SF , Ashley S, Brewster JM, Morgan VA, et al. Combined use of diffusion-weighted MRI and 1H MR spectroscopy to increase accuracy in prostate cancer detection. Am J Roentgenol. 2007;188:91-8. 36. Tanimoto A, Nakashima J, Kohno H, Shinmoto H, Kuribayashi S. Prostate cancer screening: the clinical value of diffusion-weighted imaging and dynamic MR imaging in combination with T2-weighted imaging. J Magn Reson Imaging. 2007;25:146-52. 37. Vilanova JC, Comet J, Barcelo-Vidal C, Barcelo J, López-Bonet E, Maroto A, et al. Peripheral zone prostate cancer in patients with elevated PSA levels and low free-to-total PSA ratio: detection with MR imaging and MR spectroscopy . Radiology. 2009;253:135-43. 38. Mazaheri Y, Shukla-Dave A, Hricak H, Fine SW , Zhang J, Inurrigarro G, et al. P rostate cancer: identification with combined diffusion-weighted MR imaging and 3D 1H MR spectroscopic imaging–correlation with pathologic ndings. Radiology. 2008;246:480-8. 39. Haider MA, van der Kwast TH, Tanguay J, Evans AJ, Hashmi AT, Lockwood G, et al. Combined T2-weighted and diffusion-weighted MRI for localization of prostate cancer. Am J Roentgenol. 2007;189:323-8. 40. Lim HK, Kim JK, Kim KA, Cho KS. P rostate cancer: apparent diffusion coefficient map with T2-weighted images for detection–a multireader study. Radiology. 2009;250:145-51. 41. L i H, Sugimura K, Kaji Y, Kitamura Y, Fujii M, Hara I, et al. Conventional MRI capabilities in the diagnosis of prostate cancer in the transition zone. Am J R oentgenol. 2006;186: 729-42. 42. Seitz M, Shukla-Dave A, Bjartell A, Touijer K, Sciarra A, Bastian P, et al. Functional Magnetic R esonance Imaging in P rostate Cancer. Eur Urol. 2009;55:814-91. 43. Shukla-Dave A, Hricak H, Ishill NM, Moskowitz CS, Drobnjak M, Reuter VE, et al. Correlation of MR imaging and MR spectroscopic imaging findings with K i-67, phospho-Akt, and androgen receptor expression in prostate cancer. Radiology. 2009;250: 803-12. 44. V illeirs GM, De Meerleer GO, De V isschere PJ, Fonteyne VH, Verbaeys AC, Oosterlinck W. Combined magnetic resonance imaging and spectroscopy in the assessment of high grade prostate carcinoma in patients with elevated PSA: A single-institution experience of 356 patients. Eur J Radiol. 2009. Epub ahead of print 45. K ubota Y, Kamei S, Nakano M, Ehara H, Deguchi T, Tanaka O. The potential role of prebiopsy magnetic resonance imaging combined with prostate-specific antigen density in the detection of prostate cancer. Int J Urol. 2008;15:322-6. 46. Xu S, Kruecker J, Turkbey B, Glossop N, Singh AK, Choyke P, et al. Real-time MRI-TRUS fusion for guidance of targeted prostate biopsies. Comput Aided Surg. 2008;13:255-64. 47. Shukla-Dave A, Hricak H, Eberhardt SC, Olgac S, Muruganandham M, Scardino PT, et al. Chronic prostatitis: MR imaging and 1H MR spectroscopic imaging findings–initial observations. Radiology. 2004;231:717-24. 48. Kelloff GJ, Choyke P, Coffey DS. Challenges in clinical prostate cancer: role of imaging. Am J Roentgenol. 2009;192:1455-70. 49. Engelbrecht MR, Jager GJ, Laheij RJ, Verbeek AL, van Lier HJ, Barentsz JO. Local staging of prostate cancer using magnetic resonance imaging: a meta-analysis. Eur Radiol. 2002;12: 2294-302. 50. Smith JA, Scardino PT, Resnick MI, Hernández AD, Rose SC, Egger MJ. Transrectal ultrasound versus digital rectal examination for the staging of carcinoma of the prostate:

524 results of a prospective, multi-institutional trial. J Urol. 1997; 157:902-6. 51. Claus FG, Hricak H, Hattery RR. P retreatment evaluation of prostate cancer: role of MR imaging and 1H MR spectroscopy. Radiographics. 2004;24(Suppl 1):S167-80. 52. White S, Hricak H, Forstner R, K urhanewicz J, V igneron DB, Zaloudek CJ, et al. P rostate cancer: effect of postbiopsy hemorrhage on interpretation of MR images. Radiology. 1995; 195:385-90. 53. W ang L, Mullerad M, Chen HN, Eberhardt SC, Kattan MW , Scardino PT, et al. P rostate cancer: incremental value of endorectal MR imaging ndings for prediction of extracapsular extension. Radiology. 2004;232:133-9. 54. Jager GJ, Severens JL, Thornbury JR, de la Rosette JJ, Ruijs SH, Barentsz JO. Prostate cancer staging: should MR imaging be used? A decision analytic approach. Radiology. 2000;215:445-51. 55. P oulakis V , W itzsch U, de VR, Emmerlich V , Meves M, Altmannsberger HM, et al. Preoperative neural network using combined magnetic resonance imaging variables, prostate speci c antigen and Gleason score to predict prostate cancer stage. J Urol. 2004;172:1306-10. 56. Wang L, Hricak H, Kattan MW, Schwartz LH, Eberhardt SC, Chen HN, et al. Combined endorectal and phased-array MRI in the prediction of pelvic lymph node metastasis in prostate cancer. Am J Roentgenol. 2006;186:743-8. 57. Hovels AM, Heesakkers RA, Adang EM, Barentsz JO, Jager GJ, Severens JL. Cost-effectiveness of MR lymphography for the detection of lymph node metastases in patients with prostate cancer. Radiology. 2009;252:729-36. 58. Ravizzini G, Turkbey B, Kurdziel K, Choyke PL. New horizons in prostate cancer imaging. Eur J Radiol. 2009;70:212-26. 59. Barcelo J, Vilanova JC, Riera E, Balliu E, Pelaez I, Marti J, et al. Resonancia magnética de todo el cuerpo con técnica de difusión (PET virtual) para el cribado de las metástasis óseas. Radiologia. 2007;49:407-15. 60. V enkitaraman R, Cook GJ, Dearnaley DP, Parker CC, Huddart RA, Khoo V, et al. Does magnetic resonance imaging of the spine have a role in the staging of prostate cancer? Clin Oncol (R Coll Radiol). 2009;21:39-42.

J.C. Vilanova et al 61. Cirillo S, Petracchini M, Scotti L, Gallo T, Macera A, Bona MC, et al. Endorectal magnetic resonance imaging at 1.5 Tesla to assess local recurrence following radical prostatectomy using T2-weighted and contrast-enhanced imaging. Eur Radiol. 2009;19:761-9. 62. Pucar D, Shukla-Dave A, Hricak H, Moskowitz CS, Kuroiwa K, Olgac S, et al. Prostate cancer: correlation of MR imaging and MR spectroscopy with pathologic findings after radiation therapy-initial experience. Radiology. 2005;236: 545-53. 63. Mueller-Lisse UG, Vigneron DB, Hricak H, Swanson MG, Carroll PR, Bessette A, et al. Localized prostate cancer: effect of hormone deprivation therapy measured by using combined three-dimensional 1H MR spectroscopy and MR imaging: clinicopathologic case-controlled study. Radiology. 2001;221: 380-90. 64. K urhanewicz J, V igneron DB, Hricak H, P arivar F, Nelson SJ, Shinohara K, et al. P rostate cancer: metabolic response to cryosurgery as detected with 3D H-1 MR spectroscopic imaging. Radiology. 1996;200:489-96. 65. Donnelly SE, Donnelly BJ, Saliken JC, Raber EL, V ellet AD. Prostate cancer: gadolinium-enhanced MR imaging at 3 weeks compared with needle biopsy at 6 months after cryoablation. Radiology. 2004;232:830-3. 66. Kirkham AP, Emberton M, Hoh IM, Illing RO, Freeman AA, Allen C. MR imaging of prostate after treatment with high-intensity focused ultrasound. Radiology. 2008;246:833-44. 67. Akin O, Hricak H. Imaging of prostate cancer. Radiol Clin North Am. 2007;45:207-22. 68. Sciarra A, P anebianco V, Salciccia S, Osimani M, Lisi D, Ciccariello M, et al. R ole of dynamic contrast-enhanced magnetic resonance (MR) imaging and proton MR spectroscopic imaging in the detection of local recurrence after radical prostatectomy for prostate cancer . Eur Urol. 2008;54: 589-600. 69. Sella T, Schwartz LH, Swindle PW, Onyebuchi CN, Scardino PT, Scher HI, et al. Suspected local recurrence after radical prostatectomy: endorectal coil MR imaging. Radiology . 2004;231:379-85.