Detection of extracapsular extension (ECE) is quite important in preoperative staging since its presence upstages the patient to T3A and alters the treatment to a more aggressive approach. On T2W images, ECE usually appears as a direct extension of the tumor into the hyperintense peri-prostatic fat tissue; however, frank ECE may not be seen in all situations. Under those circumstances, the secondary findings should be evaluated; these include asymmetry of the neurovascular bundle, envelopment of the neurovascular bundle, contour angulation, irregular gland margin, capsular obscuration or retraction, and obliteration of the rectoprostatic angle [8, 9]. Seminal vesicle invasion (SVI) can be directly visualized as an extension of tumor from the base of the prostate into the seminal vesicles and the presence of focal low-signal intensity within the normally ­hyperintense seminal vesicles [10]. Multiplanar evaluation of the T2W images enables more accurate evaluation of ECE and seminal vesicle invasion.

T2W MRI alone is reported to have a wide range of sensitivity and specificity for cancer detection of 27–100% and 32–99%, respectively, depending on many factors discussed below. Sensitivity and specificity for local staging and predicting the presence of extracapsular extension also demonstrate a wide range from 14.4 to 100% and 67 to 100%, respectively [11–25]. The wide range of sensitivities and specificities is mainly due to the significant variability in the patient populations, magnet strengths, coil designs as well as differences in the gold standard correlation methodologies (biopsy vs. surgery), and level of expertise used in different studies.

For a complete local staging, the status of nodal disease is crucial. However, nodal staging is challenging and depends on morphologic enlargement of the node and qualitative assessment of the enhancement pattern if intravenous contrast is used. Generally, MRI is insensitive for node involvement because it depends on node enlargement which occurs late in the natural history of advanced disease. Contrast-enhanced MRI, lymphotrophic ultrasmall superparamagnetic iron oxide (USPIO)-enhanced T2*W gradient-refocused echo MRI offers an alternative to traditional size-based lymph node evaluation [26–28]. After intravenous administration, this contrast agent is carried to lymph nodes where it is processed by dendritic cells. Benign lymph nodes can accumulate this contrast agent, whereas malignant lymph nodes, lacking macrophages, are unable to take the agent up. This allows earlier detection of nodal involvement even if the overall dimensions of the affected lymph node are of normal size [29]. The original iron-based contrast agent (Combidex^®) is no longer available; however, another similar agent carboxymethyl dextran-based magnetic nanoparticle (ferumoxytol, tradename: Ferheme^®) is available in the USA as an iron replacement therapy in kidney failure patients. Ferumoxytol can be injected as a bolus unlike Combidex^® which required infusion over 30 min. It has been reported to be safe for lymph node staging in prostate cancer patients [30], but little data about its efficacy are available.

Diffusion weighted MRI (DW-MRI) evaluates the Brownian motion of free water within tissues. It has been in use for evaluation of acute stroke patients in the last decade and recently there is growing interest in its use for the depiction and follow-up of prostate tumors. Cellularity increases in most cancer types and tissue water diffusion, thus, becomes more restricted since the mean water path length is interrupted by cell membranes. This results in reduced diffusion of water. Similar to other cancer types, prostate cancer often includes tightly packed glandular elements with increased cellularity, abundant stroma, and diminished extracellular spaces which reduce diffusion. DW-MRI is quantified by calculating the apparent diffusion coefficient (ADC) and displayed as a parametric map reflecting the relative ADC. DW-MRI is a time-efficient MR imaging technique and does not require exogenous contrast media. Prostate cancers demonstrate hyperintense signal characteristics when compared to healthy PZ on raw high “b” field DW-MR images, whereas on ADC maps they show decreased signal intensity relative to healthy PZ, reflecting lower water diffusion (Fig. 16.2) [31]. DW-MRI demonstrates a wide range of sensitivity and specificity (57–93.3% and 57–100%, respectively) for tumor detection in various studies, depending on field strength, imaging parameters, patient selection, and validation [32–38]. Additional studies have also demonstrated differences in ADC values among different Gleason grades, thus helping to stratify low-risk from higher-risk prostate cancers [39, 40]. DW-MRI is an intrinsically low SNR imaging sequence, but the emergence of high field strength magnets, the widespread use of endorectal coils filled with inert material such as perfluorocarbon, and parallel imaging approaches have yielded better SNR and spatial/temporal resolution for DW-MRI. Additionally, use of higher gradient strengths (i.e., b values of 1,000–2,000 s/mm²) improves performance for lesion detection [33, 41–43]. Air in the rectum can lead to susceptibility artifacts on DW-MRI and thus studies obtained without endorectal coils or those in which the endorectal coil is filled with air will have suboptimal DW-MRI scans.

Proton MR spectroscopy imaging (MRSI) provides a display of the chemical composition of the prostate gland through specific metabolites, such as citrate, choline, and creatine which have characteristic resonance frequencies that can be measured with MRSI. The normal prostate gland contains low levels of choline and high levels of citrate, whereas prostate cancers demonstrate increased levels of choline and diminished levels of citrate. The high choline level within prostate cancers is primarily related to increased cell turnover. There is an increased amount of soluble, free choline due to overexpression of choline kinase in prostate cancer tissue [44, 45]. Normal secretory epithelial cells of the prostate gland produce high levels of citrate, and normal prostate glandular cells possess excess zinc with the highest levels the body. Zinc inhibits the citrate oxidizing enzyme “aconitase” and blocks the Krebs cycle, which results in accumulation of citrate. However, in prostate cancer cells, levels of zinc are depleted resulting in unblocked aconitase activity, leading to further oxidation of citrate and thus depleted levels in prostate cancer cells [46, 47]. Owing to the changes in the metabolite levels within prostate cancer lesions, the ratio of choline to citrate (cho/cit) can be used as an index for detecting malignancy (Fig. 16.3). During MRSI, signal from lipid and water must be suppressed in order to focus on the signals coming from the key metabolites, choline, citrate, and creatine which resonate at 3.2, 2.6, and 3.0 parts per million (ppm), respectively. If left unsuppressed, the signal from water and lipid would overwhelm the relatively minor peaks of these key metabolites. The ­concentration of these metabolites is estimated by measuring the area under each peak after baseline corrections for water and lipid.

Currently, the majority of the experience on prostate MRSI has been at 1.5 T magnets; however, using 3 T or higher field magnets has several advantages for MRSI. Increased field strength enables smaller voxel sizes, improved temporal resolution, and more accurate separation of metabolite peaks (e.g., better separation of creatine and choline peaks). Recent advances in image acquisition and data processing software have led to consideration of other metabolites such as polyamines which are also implicated in prostate cancer metabolism [48]. Integration of MRSI into routine prostate MRI practice improves tumor detection rates [49–51] and estimation of tumor volume, extracapsular extension, radiotherapy response, and post-radiotherapy recurrence [52–54]. However, despite its benefits, MRSI is technically challenging to run since it usually requires the assistance of an onsite MRSI physicist and it is time consuming (∼13–18 min for multi-voxel MRS applications). Moreover, given the long scan duration, patient motion can be a problem. Currently there is no consensus on whether the cho/cit or the cho+cre/cit ratio should be used to analyze the data and what criteria are employed for positivity (generally the ratio should be two deviations above the mean normal ratio) [55, 56].

Angiogenesis is a well-known property of tumor growth. Neoangiogenic tumor vessels are anarchic, inefficient, and highly permeable that are distinct in morphology and overall architecture compared with normal vasculature. The enhancement pattern of a neoplasm depends on the richness of its blood supply and its mean capillary permeability. DCE-MRI is performed with fast T1 weighted MR imaging sequences (mostly gradient recalled echo-GRE) before, during, and after the rapid bolus administration (2–4 cc/s) of a low molecular weight gadolinium chelate. Region of interest measurements over the tumor generate enhancement curves that can be analyzed subjectively or mathematically fit to a two compartment pharmacokinetic model. Tumors show early intense enhancement compared to normal tissue. Moreover, tumors more rapidly de-enhance or washout than normal tissue. The pharmacokinetic model produces several transport parameters, such as K ^trans (transendothelial transport of contrast medium from vascular compartment to the tumor interstitium), k _ep (reverse transport parameter of contrast medium from the extracellular space back to the plasma space), fpV (fraction of plasma volume compared to whole tissue volume) and V _e [extravascular, extracellular volume fraction of the tumor (the fraction of tumor volume occupied by EES)] for describing tumor and tissue permeability [57]. Higher grade tumors tend to have higher K ^trans and k _ep parameters compared to normal surrounding prostate tissue (Fig. 16.4). Abnormal enhancement patterns are seen in both tumor foci and BPH nodules, making assessment of the central gland difficult [58, 59]. DCE-MRI alone has reported sensitivity and specificity ranges of 46–96% and 74–96%, respectively, for detection of tumor lesions but, as always, these ranges are highly dependent on patient selection, technique, and diagnostic criteria [32, 60–64]. Despite its accepted utility for detection of primary prostate cancer and posttreatment recurrence, some challenges exist. The parameters generated by the two compartment model have high standard deviations within a tumor and between patients. Moreover, inter-institutional variability is even greater due to variations in implementation. Standardization of image acquisition protocols and analysis techniques is needed to overcome this problem. There is also now new concern over inducing nephrogenic systemic fibrosis (a severe interstitial fibrosis that occurs after intravenous injection of gadolinium chelates) in patients with severe renal failure, especially those undergoing dialysis; therefore, the use of gadolinium-contrast agents should be carefully evaluated in such patients.