Recent advancements in CT and MR imaging of Prostate Cancer

Abstract

Computed tomography (CT) and magnetic resonance imaging (MRI) are both commonly used in prostate cancer (PCa) management, which includes a large spectrum from screening positive pre-diagnosis phase to metastatic disease. CT and MRI have continually evolved to meet the changing demands for PCa management. For CT, novel techniques such as dual energy CT and photon counting CT show promising results for tissue characterization and quantification. For MRI, the detection, staging, and management of prostate cancer has been significantly improved by the development of multiparametric, biparametric, and whole-body MRI techniques. Additionally, research on ultrasmall superparamagnetic particles of iron oxide (USPIO) contrast-enhanced MRI has revealed promising results for nodal staging of PCa. In this manuscript we aim to outline the current status and recent advancements of CT and MRI in PCa imaging.

Introduction:

Computed tomography (CT) and magnetic resonance imaging (MRI), both first introduced into medicine in the late 1970s and 80s, completely revolutionized the field of medicine (1, 2). Both modalities allowed for better diagnosis, treatment, and overall patient outcomes compared to the existing imaging technologies (1, 2). In clinical management of prostate cancer (PCa), both CT and MRI have been widely used for initial diagnosis and staging purposes. The role of MRI and CT in PCa has evolved significantly as new technologies become available. Ranging from novel CT techniques to shortened MRI protocols and unique contrast agents, the field continues to explore methods to improve image acquisition, cancer diagnosis, and management. In this manuscript we aim to outline the current status and recent advancements of CT and MRI in PCa imaging.

Computed Tomography (CT)

Current status:

CT has a limited role in the diagnosis of primary PCa and currently it is mainly used for distant staging in newly diagnosed PCa patients, or to evaluate lymph node and bone metastases during the course of metastatic PCa (3, 4). Although it is a commonly used modality in PCa management, CT imaging does not provide sufficient soft tissue contrast and lacks targeted molecular information (5).

Along with imaging limitations, interpretation of CT scans for nodal staging of PCa is entirely reliant on size-based morphologic criteria. For lymph node evaluation, routine CT protocols are limited by size criteria-small nodes (1.0 cm in short axis of oval nodes and 0.8 cm for round nodes) for indication of metastatic disease (6). However, more than half of metastatic PCa lymph nodes may be less than 1cm, while others are enlarged due to non-cancer reasons, such as reactive hyperplasia (6). A meta-analysis by Hövels et al. found that for lymph node metastasis detection, CT had a pooled sensitivity of 42% (95% CI, 26–56%) and a pooled specificity of 82% (95% CI, 79–83%) (7). For bone evaluation, metastatic PCa lesions typically appear as osteoblastic foci in the bone marrow and CT can robustly reveal bone metastases, however, CT may not always be able to show early phase bone metastases, which is why PET techniques with novel tracers have been developed and whole-body MRI (WB-MRI) has been explored. In general, CT has insufficient sensitivity to accurately detect PCa unless it reaches a certain size and more molecularly targeted approaches utilizing positron emission tomography (PET) in combination with CT has been investigated for PCa diagnosis for the last two decades. In a study by Hofman et al. of 302 patients, prostate-specific membrane antigen (PSMA) PET-CT was found to have superior accuracy and improved sensitivity and specificity compared to conventional imaging for the detection of pelvic nodal or distant metastatic disease (8). Now that, there are there are approved PET tracers for PSMA imaging, use of PSMA targeting PET/CT is expected to increase gradually in near future (9).

Advancements in CT Technology

Dual-energy CT:

Dual-energy CT (DECT) is based on the use of two different photon energy levels to identify and quantify the composition of various materials, including fat, calcium, and iodine. The difference in attenuation allows for tissue characterization as well as quantification of each material to create concentration maps. This novel approach is documented to allow for improved lesion detection and characterization (10). DECT measurements are based on algorithms from the attenuation data, therefore, a more objective and quantitative evaluation is obtained compared to MRI (11). DECT does not expose patients to extra radiation that is significantly higher than standard single-energy CT (12). The improved tissue and material characterization and comparable (or smaller radiation) dose make DECT a better imaging technique than standard CT.

Currently, there is not a widely reported use of DECT in clinical management of PCa; however, the research in this field is continuing. To evaluate the improvement in tissue characterization using DECT in the prostate, a phantom model was created from a patient’s pelvic CT scan. Monte Carle dose calculations for lose-dose rate prostate brachytherapy were performed. DECT images were analyzed with the Bayesian eigentissue decomposition method. While the DECT method appeared to be more sensitive to noise than the single-energy CT (SECT), the DECT performance at all noise levels was equivalent with and without calcification, implying it is robust even with noise (13). DECT gains over SECT were more noticeable with calcifications, and DECT also provided dose distribution with the highest accuracy and low noise, particularly for organs at risk (13). Although a phantom study, the results of this work indicated advantages of use of DECT in radiation treatment management of PCa.

DECT is a novel CT technology and although its use in PCa is not well documented, it is important to be aware of pitfalls and artifacts that may impact the image quality of DECT images. One pitfall is patient weight and its impact on material decomposition and tissue characterization. A consensus panel has a recommended weight (260–280 lb. [118–127 kg]) and transverse diameter (38–46 cm) range of cutoff values for selecting patients for which DECT in the abdomen would provide high-quality images (14). Acquisition parameters may also be modified to improve image quality and the potential benefits of DECT imaging in such populations (14).

Iodine “blooming,” wherein iodinated contrast medium has significant amplification on low-keV and iodine images, can cause images to be streaky which may impact visualization of disease or bowel folds (14). While changing the window parameters may mitigate artifacts, it is often necessary to reduce contrast medium concentration (14). This reduction will reduce contrast-related streaking without hindering diagnostic performance and may improve the quality of virtual non-contrast (VNC) images (14). The field of view (FOV) in DECT is smaller than in standard CT so technologists should be mindful of patient positioning.

Photon Counting CT:

While a standard CT uses an energy-integrating detector (EIDs) to register the total x-ray energy deposited per interval, photon-counting detectors (PCDs) can register the x-ray energy deposited per photon which allows for a spectral image to be created. This results in a higher contrast-to-noise ratio (CNR), improved spatial resolution, and reduced radiation exposure compared to a standard CT (15). PCD-CT also allows for the ability to image with and differentiate between multiple contrast agents (16).

There are a few challenges with PCDs. One challenge is the cross talk which occurs when photons hit one detector and then scatter, hitting a neighboring detector. This results in blurry images, extra noise, and a degraded energy resolution because photos have been counted more than once (15). Another challenge is pulse pile-up. When photos arrive to the detector too quickly, they end up overlapping which causes them to register as a single pulse with larger energy. The image on impact quality is an increase in noise as well as energy resolution deterioration, however, the degradation is not equal across the image (15). Like DECT, the spectral data from PCD-CT can be used to generate virtual monoenergetic images (VMI) which can be used to salvage poor scans.

While the technique holds great promise for clinical use, there is no documented research or clinical experience on use of PCD in PCa. Further research into the development of PDC-CT technology is needed to reduce the cost of commercialization. In the meantime, further research on overcoming technical techniques will help refine the acquisition process for future use.

Summary:

In summary, CT imaging is traditionally known to be limited in PCa management since it is limited to size-based morphologic evaluations, and it lacks tissue and molecular characterization. Both DECT and PCT are novel methods which enable tissue characterization and quantification. DECT allows for a more objective evaluation of in vivo tissues compared to MRI and there is no significant increase in radiation exposure. There are still many technical issues related to DECT, and pitfalls when it comes to imaging on patients. Although not in routine use in PCa yet, there is an open clinical trial which aims to establish a more accurate and precise way to image metastatic bone disease in patients with prostate cancer for staging and monitoring response to therapy in Duke University Medical Center. This trial was opened in 2017 for accrual and it is currently suspended for accrual due to budgetary issues (https://clinicaltrials.gov/ct2/show/study/NCT03111914 Accessed 10/29/2021).

Photon counting CT has been found to improve CNR and spatial resolution while reducing radiation exposure compared to a standard CT. It creates opportunities for quantitative imaging but there are too many technical challenges that result in degraded image quality for it to be widely used. The cost of the technology has been prohibitively high until recently, in upcoming years we can expect research in PCD-CT to flourish. PCD-CT is on the verge of becoming clinically feasible but has not yet been perfected. Neither of these novel CT techniques is ready for widespread clinical use for PCa, but both promise great improvement in performance and diagnostic use in the future.

Magnetic Resonance Imaging (MRI)

MRI is a technique that uses powerful magnets to align hydrogen atoms and radio pulses to detect proton orientation, density, and diffusion in the tissue. This information is used to produce cross-sectional images of the body. Gadolinium based intravenous (IV) contrast agents may be used to improve tissue contrast, but MRI can aid tissue characterization without use of IV contrast. MRI stands as an ideal imaging technique for localized PCa imaging because it is non-invasive, has better spatial resolution, strong soft tissue contrast and multi-planar image acquisition capability (17). Various protocols have been developed to improve the resolution of images, reduce acquisition time, and further explore optimal use of MRI technology. Currently, MRI is best used in the early detection of PCa, biopsy guidance, local staging, and active surveillance (18).

Novel MRI Techniques

Multiparametric MRI

Multi-parametric MRI (mpMRI) allows for a superior soft tissue resolution for localized PCa imaging by combining anatomic and functional pulse sequences which results in increased sensitivity predictive value for cancer detection (17). Specifically, combining T1 weighted (T1W) images, T2 weighted (T2W) images, and diffusion weighted images (DWI) with apparent diffusion coefficient (ADC) maps and dynamic contrast enhanced (DCE) images allow for diagnostic improvement. For the anatomic pulse sequences, T1W images provide information on the presence of hemorrhage or scarring following biopsy. Such iatrogenic impacts on the prostate may obscure the visualization of cancers; despite a delay of 6 to 8 weeks after biopsy, hemorrhage may still be present, in which case the examination should be reschedule (18). T2W images provide information on the zonal architecture of the prostate. The peripheral zone, transition zone, prostatic urethra, prostatic capsule, and seminal vesicles are well defined (18). This definition helps radiologists identify cancer suspicious lesions, which appear more hypointense than the healthy peripheral zone tissue (18). Benign prostatic hyperplasia (BPH) is a commonly occurring process in the transition zone in aging individuals and MRI findings of PCa can sometimes overlap with those of BPH; however, PCa lesions usually appear hypointense with irregular margins or lenticular shape in the transition zone. Such morphologic features can be quite useful to detect PCa lesions. T2W MRI not only provides accurate anatomic information to detect PCa lesions, but it also provides valuable information for staging purposes such as relationship of PCa lesions with prostatic capsule, seminal vesicles, neurovascular bundles and surrounding critical organs (e.g., rectum, bladder).

The diagnostic accuracy of T2W MRI is improved by the functional MRI sequences DWI and DCE MRI (figure 1). DWI MRI reflects the Brownian motion of water molecules within the prostatic tissue. Cancer suspicious areas, which include highly dense cellular structure have lower water diffusion compared to the normal prostatic tissue. DWI MRI can be robustly quantified by using apparent diffusion coefficient (ADC) maps using different types of signal decay models. The most commonly used one is the mono-exponential decay model. The ADC map is comprised of DWI MRI acquired with different magnetic gradient strengths, called “b-values”, the ADC value of each voxel obtained reflects the diffusivity of water in the prostate. In PCa lesions, water diffusivity is restricted resulting in a lower ADC value compared to the normal tissue. ADC values are not only helpful in detecting cancers but can also help to predict PCa lesions with higher biologic aggressiveness (19). Another important vital component of DWI MRI is the high b-value DWI. At high (≥1500 s/mm²) b-value MRI, the non-cancerous prostate tissue is expected to have less signal due to increased water diffusivity compared to PCa tissue. Although high b-value DWI MRI is a non-quantifiable pulse sequence, PCa can appear prominently hyperintense and readily visible which is quite critical for diagnosis purposes (18).

Figure 1:

65-year-old male with a serum PSA of 7.7ng/ml. Axial T2W MRI (a), ADC map (b), b1500 DW MRI (c) and DCE MRI (d) shows a focal lesion in the right apical peripheral zone (arrows). Based on multiparametric MRI evaluation, this lesion was interpreted as a PI-RADS 4 category and TRUS/MRI fusion guided biopsy revealed Gleason 4+4 prostate adenocarcinoma.

DCE MRI is another functional pulse sequence which aims to detect tumor related angiogenesis of PCa. DCE MRI is a fast gradient echo pulse sequence which is obtained before, during, and after the IV injection of gadolinium-based contrast. DCE MRI theoretically allows for rapid wash in and wash out of contrast material within the prostate. PCa lesions typically display focal and early contrast enhancement compared to normal prostate tissue. However, DCE MR findings of PCa can overlap with normal conditions, such as benign prostatic hyperplasia, prostatitis, atrophy, or prostatic intraepithelial neoplasia which warrants its combined evaluation with T2W and DWI MRI (18). Currently, there is a debate on the utility of DCE MRI in PCa detection. The need for the incremental gain brought by DCE MRI in the detection of PCa when T2W and DWI MRI are already obtained is unclear. Additionally, DCE MRI adds to the cost and invasiveness of the examination (18).

mpMRI has been documented to improve PCa diagnosis. In one prospective study, to determine the PCa detection rate using mpMRI, 45 biopsy-proven PCa patients with a mean prostate specific-antigen (PSA) of 6.37ng/mL underwent a prostate MRI prior to prostatectomy (17, 20). The MRI imaging protocol included triplanar T2W turbo spin echo (TSE), DWI MRI, 3D MR Point Resolved Spectroscopy, DCE MRI sequences. The mpMRI sequences significantly improved the positive predictive value for the overall prostate, peripheral zone, and transition zone (17). The positive predictive value (PPV) for detection of PCa using mpMRI was 98%, 98%, and 100% in the overall prostate, peripheral zone, and transition zone, respectively (17). The sensitivity of the mpMRI sequences was higher for tumors >5mm in diameter and for tumors with Gleason scores >7 (p<0.05) (17).

In several clinical studies mpMRI was reported to improve PCa detection. This has expanded the use of mpMRI for guiding prostate biopsies, which refer to MRI guided prostate biopsy. The MRI guided prostate biopsy approach can be done using different strategies such as cognitive, in-bore and TRUS/MRI fusion guided methods (21).

Use and utility of mpMRI in guiding prostate biopsies has been reported in multiple studies. A multicenter trial was conducted with 500 patients that had a clinical suspicion of PCa to determine if mpMRI, with or without targeted biopsy, was an alternative to TRUS biopsy for PCa detection. The patients had elevated PSA levels and were biopsy naïve. A total of 500 patients were randomly split into one group of 252 for MRI-targeted biopsy and another group of 248 for standard-biopsy. In the MRI-targeted biopsy group, 95 (38%) patients were found to have clinically significant PCa versus 64 (26%) such patients from the standard-biopsy group (P=0.005)(22). This level 1 evidence reveals that mpMRI, with or without, biopsy is noninferior to the standard TRUS-guided biopsy (22). Another prospective study of 1003 patients undergoing targeted and standard biopsy from 2007 through 2014 found that targeted MR/ultrasound fusion biopsy was more accurate than standard US-guided biopsy (23). 30% more high-risk cancer were diagnosed by targeted biopsy vs standard biopsy (173 vs 122 cases, P < .001) (23). Targeted biopsy also diagnosed 17% fewer low-risk cancers (213 vs 258 cases, P < .001), and when standard biopsy cores were combined with targeted cores, an additional 103 cases (22%) of mainly low-risk PCa was diagnosed (83% low risk, 12% intermediate risk, and 5% high risk) (23). The mpMRI can guide needle biopsies more accurately compared to systematic biopsy methods and diagnoses fewer low-risk PCa while diagnosing more clinically significant PCa. It also provides the basis for image-guided, organ sparing focal treatment of PCa that is minimally invasive compared to whole gland treatment approaches such as surgery and radiotherapy.

Because information provided by mpMRI can be utilized for many different procedures and informs clinicians on a wide variety of topics, its use has been increased in routine practice. Based on a retrospective review of an institutional clinical database of four million patients, Oberlin et al. found that 1521 mpMRIs were performed with an increase in use of 486% over the 26-month period of data collection (October 2013 to December 2015) (24). Prostate mpMRI is increasingly used to diagnose and manage PCa and has proven its ability to assist image guided biopsy in real time (18). mpMRI can fulfill the need for many roles in localized PCa management, including evaluation after negative biopsy, planning for surgery, biochemical recurrence, and active surveillance. The meaningful clinical information from mpMRI combined with an increasing awareness of the risks of prostate biopsy allows urologists to spare patients from invasive procedures while still effectively stratifying the risk of PCa. The successful evolution of MRI to meet growing needs in the field of prostate cancer has undoubtedly changed millions of lives for the better.

Biparametric MRI

The large cost and lengthy time expense of a mpMRI still pose as a barrier for its wider use. Additionally, mpMRI can be considered as an invasive method due to the use of Gadolinium based contrast. While Gadolinium-based contrast agents have been proven to be effective and safe in most patients, some of these contrast agents have a small risk of negative effects such as nephrogenic systemic fibrosis (NSF). NSF is an untreatable condition that is linked to Gadolinium contrast exposure during MRI (25). Considering the risks vs. benefits of DCE MRI has led to increased use of abbreviated or biparametric MRI (bpMRI) in PCa imaging. A bpMRI involves only a T2W image and DWI sequences, making it faster to obtain and interpret images (22) (figure 2). The elimination of the DCE MRI sequence may not significantly impact the diagnostic accuracy of bpMRI, and it also eliminates the risks related with IV contrast injection (22).

Figure 2:

75-year-old male with a serum PSA of 27.1ng/ml. Axial T2W MRI (a), ADC map (b) and b1500 DW MRI (c) shows a focal lesion in the left apical transition zone (arrows). Based on biparametric MRI evaluation this lesion was interpreted as a PI-RADS 4 category and TRUS/MRI fusion guided biopsy revealed Gleason 3+4 prostate adenocarcinoma.

Recently, performance of bpMRI for PCa detection has been reported in the literature. In one study, to determine the diagnostic accuracy of bpMRI in identifying significant PCa and ruling out insignificant PCa, 1020 biopsy-naïve patients with suspicion for PCa were imaged prior to biopsy. In this prospective study, Boesen et al. found that significant PCa diagnoses were improved by 11% while insignificant diagnoses were reduced by 40% (26). Restricting standard and mpMRI guided biopsy to the patients with suspicious bpMRI findings allowed 30% of patients to avoid a biopsy. The negative predictive value in ruling out significant cancer after bpMRI was 97% (26). The results of this indicated that a bpMRI can be performed as part of a triage test to quickly and effectively detect or rule out significant PCa.

Clinical data, such as PSA level, can be combined with imaging results to optimize PCa diagnosis and management when bpMRI is used. The diagnostic value of bpMRI in conjunction with PSA-based detection of PCa in biopsy naïve patients was evaluated in 143 patients (27). The bpMRI findings were assessed along with PSA and PSA density (PSAD) to screen for association with PCa detection on biopsy. bpMRI combined with PSA level led to a sensitivity of 90% while bpMRI combined with PSAD led to a specificity of 86% (27). The negative predictive value for PSAD with bpMRI was 70% was for PSA with bpMRI it was 82% (27). Combined use of bpMRI with PSA or PSAD has a greater benefit versus any single modality. For patients with challenges placing an endorectal coil and/or the use of gadolinium-based contrast agents which require IV access, bpMRI provides a potential adjunct tool to improve PCa detection and management. bpMRI also reduces the time the study requires which may be significant for centers with many patients and a great need to quickly stratify risk.

In a meta-analysis, ten studies with 1,705 patients with 3,419 lesions were evaluated to determine the diagnostic accuracy of bpMRI versus a standard mpMRI for PCa. Guided biopsy or prostatectomy histopathology results were used as ground truth. Kang et al.’s meta-analysis found no statistically significant difference in the diagnosis of PCa between bpMRI and mpMRI (28). The results also showed that there was no statistically significant difference in both sensitivity and specificity between the two (28). Despite some heterogeneity in the diagnosis cutoff value, this meta-analysis shows that bpMRI for PCa diagnosis is similarly accurate to mpMRI and may replace the time-consuming protocol. These results support the use of bpMRI for reducing image acquisition time and interpretation time while retaining good diagnostic accuracy. Similar to mpMRI, bpMRI can also be helpful in patient populations with challenging clinical scenarios. In patients who had persistent PCa suspicion despite negative biopsies, bpMRI continued to display comparable diagnostic accuracy to mpMRI. A study conducted in 542 patients with elevated PSA levels (PSA greater than or equal to 3 ng/mL) and a negative TRUS biopsy aimed to determine the diagnostic accuracy of bpMRI for clinically significant PCa. This study found that the diagnostic accuracy for bpMRI was 89.1% (483 of 542) which is nearly identical to that of full mpMRI at 87.2% (473 of 542) (29). The advantage is that the bpMRI takes only 9 minutes, does not involve contrast, and offers an equivalent cancer detection rate (25.5% in bpMRI vs 25.6% in mpMRI) (29). This kind of improvement allowed for minimally invasive detection of significant PCa even in patients where a TRUS guided biopsy did not detect it. The exact improvement in long-term care, cost, time, and application is pending but holds great promise for patients and all involved healthcare workers. Overall, experience in the literature in bpMRI is growing and further research will aid to establish its use in different clinical scenarios of localized PCa management.

Whole-body MRI

A whole-body MRI (WB-MRI) is an imaging modality which scans the patient’s entire body from head to toe to detect cancer lesions. WB-MRI protocol commonly includes a combination of anatomic T1W and T2W MRI with functional DWI MRI. For PCa patients it is currently used to detect and monitor nodal and bone metastasis.

Use of WB-MRI is relatively newly implemented for PCa imaging since acquisition of DWI MRI has become available for whole body MRI scanning recently (30). In one study, to determine patients with lymph node metastasis and oligometastatic disease, WB-MRI with DWI was performed in two groups of patients. The WB-MRI consisted of “head-to-toe” imaging and the DWI sequences were “head-to-thigh.” The first group is a newly diagnosed, untreated, asymptomatic metastatic prostate (mHNPC) cancer cohort of 46 patients who were found to have bone or lymph node metastasis on follow-up imaging. The second group was made of 50 patients who were transitioning from non-metastatic to metastatic disease (mCRPC) with a doubling of PSA in less than 6 months. Bone metastases were found in 33 of 46 patients (71.7%) in the mHNPC group and 33 of 50 (66%) in the mCRPC group, predominantly in the spine and pelvis (31). Enlarged lymph nodes, confirmed as metastatic by follow up, were found in 32 of 46 patients (69.6%) of the mHNPC group and 34 of 50 (68%) of the mCRPC group (31). 28% of the mHNPC and 52% of mCRPC patients had oligometastatic disease with more than two-thirds of the metastatic patients having lymph node metastases outside of the recommended targets for extended lymph node dissection and pelvic external beam radiation therapy (31). The results of this study indicated that WB-MRI can be ideal in that it minimizes radiation exposure and is affordable. It is also able to detect bone and lymph node metastases outside of the standard targets, which can improve PCa management.

While WB-MRI can be helpful for PCa management, it requires a great level of skill to read and awareness of common pitfalls across all areas of the body. Analysis of interobserver agreement of 50 WB-MRI examinations performed between December of 2016 and February of 2018 was retrospectively studied. A senior radiologist with nine years of experience and a resident radiologist with six months of training used MET-RADS-P guidelines to report on the detection of metastatic diseases in 14 body regions. By Cohen’s Kappa statistics, the highest interobserver agreement in the primary response assessment categories (RAC) was highest in the cervical, dorsal, and lumbosacral spine, pelvis, limbs, and lung sites (K: 0.81 – 1.0) (32). It was only moderate in the pelvic nodes (K: 0.56) but for secondary RAC it was excellent in the cervical spine and retroperitoneal nodes (K: 0.93, 0.89, respectively) (32). The clinical need for monitoring metastasis in patients who receive a WB-MRI is growing, so the need for quality interobserver agreement is a priority. This set of guidelines shows potential as two radiologists at very different experience levels had a high level of agreement.

Iron Oxide Enhanced MR Lymphography

Detection of nodal metastasis is quite critical for PCa management. Similar to CT scans, traditional evaluation for MRI is also heavily dependent on size based morphologic interpretations, which often results in false positive and false negative results for MRI. The need for a novel approach for MRI to identify lymph node metastasis had been addressed by iron oxide nanoprobes and particles. These particles, called ultrasmall superparamagnetic particles of iron oxide (USPIO), are biologically stable and are generally nontoxic (25). Ferumoxtran-10, a popular USPIO, is privately made in the Netherlands but is not widely commercially available.

Ferumoxtran-10 has been reported to be useful in detecting nodal metastasis in PCa in a number of studies. A study in patients with bladder and PCa using USPIO aimed to evaluate the usefulness of USPIO-enhanced MR imaging in staging of normal-sized lymph nodes. 21 patients were enrolled and a diagnostic accuracy of 90% was determined (33). The standard method of evaluating USPIO-enhanced MR images with and without DW-MRI versus histopathology took approximately 80 minutes (range 45–180 min) while the USPIO-enhanced MR imaging method took 13 minutes (range 5–90 min) (33). Of the 26 metastases found through histopathology, USPIO-enhanced MR imaging correctly diagnosed 24 of the nodes (93%) (33). Two patients had one micro-metastasis each that was missing in all imaging studies (33). Another prospective clinical trial using ferumoxtran-10 in 26 patients that were post-radical prostatectomy for the treatment of lymph node metastasis from primary PCa found promising results. Image guided biopsy was used as the reference standard. Even though this population was not expected to have grossly involved lymph nodes, the rate of positivity was 23% (34). Of the 26 patients, six who were believed to have negative nodes tested lymph node positive after imaging with USPIO-enhanced MR (34). Nine positive nodes which were not considered to be enlarged were found in these six patients (34). Ferumoxtran-10 is documented to be helpful in detecting nodal metastasis however its limited availability is a challenge, moreover, recently reported benefit of targeted positron emission tomography (PET) imaging with prostate specific membrane antigen (PSMA) targeting tracers (figure 3) and fluciclovine PET in detecting PCa metastasis is expected to diminish its routine clinical use (35, 36).

Figure 3:

69-year-old male with a serum PSA of 1.1ng/ml (S/P prostatectomy 3 years ago). Axial T2W MRI shows an 8mm left internal iliac chain lymph node (arrow) (a), which shows diffusion restriction at b900 DW MRI (black arrow) (b). 18F-DCFPyL PSMA PET/CT shows uptake within the lymph node (arrow) (c). CT guided biopsy revealed prostate adenocarcinoma within this lymph node (arrow) (d).

Ferumoxytol is another iron oxide particle which has been used for detecting nodal metastases in PCa. Ferumoxytol was developed for iron replacement in kidney failure patients however in some studies, its use has been documented for nodal staging of genitourinary cancers. To evaluate the utility of ferumoxytol in the detection of lymph node metastasis, 39 patients with prostate, bladder, or kidney cancer who underwent ferumoxytol-enhanced MR lymphography 24 and 48 hours after 7.5 mg Fe/kg of ferumoxytol was injected into the IV were retrospectively studied (37). The sensitivity, specificity, positive predictive value, and negative predictive value of lymph node based ferumoxytol-enhanced MR lymphography was 98.0%, 64.4%, 86.0%, and 93.5%, respectively (37). The size of the lymph nodes did not affect specificity or sensitivity and signal intensity between benign and malignant lymph nodes after lymphography was similar after 24 and 48 hours (37). This suggests that imaging can safely be performed up to 2 days after injection. While ferumoxytol is FDA approved and, therefore, more widely available, the potential for iron overload and repeatability of the imaging is a major concern. Further, the contrast must be administered diluted as an IV infusion over a minimum of 15 minutes to avoid major side effects and the imaging cannot take place until the day after injection. Ferumoxytol is not approved by FDA for nodal staging of PCa therefore such use can only be possible in clinical trials. Along with these difficulties and the recently documented benefit of fluciclovine PET and PSMA targeting PET tracers, iron oxide enhanced MR lymphography stands as a less preferred method in nodal staging of PCa.

Future Use of PET/MR

The inherent low anatomic spatial resolution of PET requires it to be combined with a higher resolution modality such as CT. However, MRI can provide better resolution of tissue and higher sensitivity compared to CT. Use of MRI also prevents the exposure of ionizing radiation that comes from CT. Additionally, PET/MRI offers a logistic advantage since it is a single step imaging technique which can offer simultaneous molecular, functional and anatomic imaging data. For these reasons, use PET/MRI is becoming more common for oncologic imaging. For prostate cancer, there is a growing experience on use of PSMA targeted PET/MRI in biochemical recurrence and localized high risk PCa patients. The increase in availability of PET/MRI hybrid scanners has made exploration of this imaging significantly easier. A systematic analysis of data from 2013 to 2020 found 23 studies which included 2059 patients who underwent hybrid PET/MRI imaging (38). The pooled sensitivity for staging per patient was found to be 94.9% for primary tumors and 66.7% for lymph node metastases (38). Pooled specificity for staging was higher per lesion at 90.9% for primary tumors and 97.4% for lymph node metastases (38). In another metanalysis by Wang et al. which included 707 patients found that, the pooled sensitivity and specificity of PET/MRI in detecting primary PCa were 0.83 and 0.81, respectively. Whereas for BCR, the pooled cancer detection rate was 76 % and it positively correlated with PSA levels (39). The current evidence in the literature indicates that PET/MRI has potential application for the diagnosis of primary tumor and detection of recurrent diseases in PCa.

Summary:

Multiparametric MRI is the current standard imaging modality for detection, biopsy guidance and local staging of PCa. Biparametric MRI, which is faster and simpler compared to mpMRI, is reported to be accurate enough to be used as a triage test for PCa. It spares 30% of patients for unnecessary biopsy while improving patient risk stratification. It also does not require contrast agents to be used which makes it a good choice for patients with renal failure. Current research indicates that the accuracy of bpMRI is similar to that of mpMRI however prospective studies are needed to further confirm this performance.

Whole-body MRI is useful in the detection of lymph node and bone metastasis for PCa patients. WB-MRI minimizes radiation exposure while screening a large area of the patient’s body for cancer lesions. It also allows for local staging of metastasis before and after therapy which makes it a safer and effective replacement for CT (4). PET/MRI stands as an important single step imaging modality for PCa patients specifically after recent approval of PCa targeting radiotracers.

Iron oxide enhanced MR lymphography has been documented to be useful for nodal staging, however limited availability of some of these agents (e.g., ferumoxtran-10), logistic needs such as imaging the patients on two different days (pre-injection and 24–48 hours after injection) and recently proven benefits of targeted PET imaging make use of iron oxide MR lymphography impractical choices.

Conclusion

CT and MRI have been used in the prostate cancer care for several decades with continually growing and evolving roles. CT has a well-defined role in the management of PCa. It is used to detect and stage metastatic disease to the bone and lymph nodes. While new techniques have been developed to improve their application to PCa, they are not yet ready for widespread clinical use. MRI has evolved from a tool used mainly for local detection and staging of prostate cancer to one which can be used for image guided biopsy, active surveillance. While mpMRI is becoming the new MRI standard for PCa, the conservative bpMRI approach can allow for a sufficient and accurate diagnosis as well. WB-MRI allows for the detection and staging of distant metastasis and has shown good performance in clinical trials thus far. As techniques continue to develop and evolve, future management of PCa will become stronger and more effective.