MRI:

ARTICLE IN PRESS

PET/MRI: Emerging Clinical Applications in Oncology Tyler J. Fraum, MD, Kathryn J. Fowler, MD, Jonathan McConathy, MD, PhD Abbreviations and Acronyms AC attenuation correction CHO [11C]choline CMS Centers for Medicare & Medicaid Services CT computed tomography

Positron emission tomography (PET), commonly performed in conjunction with computed tomography (CT), has revolutionized oncologic imaging. PET/CT has become the standard of care for the initial staging and assessment of treatment response for many different malignancies. Despite this success, PET/CT is often supplemented by magnetic resonance imaging (MRI), which offers superior softtissue contrast and a means of assessing cellular density with diffusion-weighted imaging. Consequently, PET/MRI, the newest clinical hybrid imaging modality, has the potential to provide added value over PET/CT or MRI alone. The purpose of this article is to provide a comprehensive review of the current body of literature pertaining to the clinical performance of PET/MRI, with the aim of summarizing current evidence and identifying gaps in knowledge to direct clinical expansion and future research. Multiple example cases are also provided to illustrate the central findings of these publications. Key Words: PET/MRI; PET/MR; Oncology; mMR; Whole-body imaging. © 2015 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.

DWI diffusion-weighted imaging FACBC anti-1-amino3[18F]fluorocyclobutyl-1carboxylic acid FDG 2-deoxy-2-[18F]fluoro-Dglucose FDOPA 6-[18F]fluoro-3,4-dihydroxyphenylalanine FET O-(2-[18F]fluoro-ethyl)-Ltyrosine FMISO [18F]fluoromisonidazole MET L-[11C]methionine MRI magnetic resonance imaging OMs osseous metastases PET positron emission tomography

Acad Radiol 2015; ■:■■–■■ From the Mallinckrodt Institute of Radiology, Washington University, Campus Box 8131, 510 S. Kingshighway Blvd., Saint Louis, MO, 63110. Received August 8, 2015; revised August 8, 2015; accepted September 27, 2015. Funding acknowledgements: Support for the [18F]FDOPA-PET/MRI study was supported through grant K08CA154790 from the National Cancer Institute and by Grant #ACS IRG-58-010-55 from the American Cancer Society. Address correspondence to: T.J.F. e-mail: [email protected] © 2015 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.acra.2015.09.008

1

FRAUM ET AL

Academic Radiology, Vol ■, No , ■■ 2015

PSA prostate-specific antigen PSMA prostate-specific membrane antigen SCC squamous cell carcinoma SUV standardized uptake value SUVmax maximum SUV TSE turbo spin echo

INTRODUCTION

P

ositron emission tomography (PET) has revolutionized the imaging evaluation of numerous oncologic conditions by exploiting biochemical and physiologic differences between tumor cells and normal tissues (1). Often performed in conjunction with computed tomography (CT), PET utilizing the glucose analog 2-deoxy-2-[18F]fluoro-Dglucose (FDG) has become the standard of care for the initial staging and the subsequent assessment of treatment response for many malignancies (2,3). Tumor uptake of FDG reflects the increased rates of aerobic glycolysis that occur in many cancer cells (the Warburg effect) relative to most normal tissues and benign lesions. The resulting distribution of FDG thereby allows for anatomic delineation of local and distant tumor spread by PET/CT and provides a measure of a key aspect of cancer metabolism. Many PET tracers have also been developed to take advantage of other distinctive tumor properties, such as elevated amino acid transport or altered receptor expression (4). Despite its proven utility, FDG-PET/CT has important limitations, especially with respect to local tumor staging and the characterization of certain incidental lesions. In such situations, further evaluation with magnetic resonance imaging (MRI) may be indicated to achieve optimal clinical management. The superb soft-tissue contrast of MRI and its capacity to assess cellular density by diffusion-weighted imaging (DWI) constitute powerful supplements to the molecular and metabolic data of PET. Consequently, PET/MRI, the newest clinical hybrid imaging modality, has significant potential to improve the diagnosis, initial staging, and subsequent restaging of numerous cancers. However, studies demonstrating such benefits are needed to support the routine clinical use of PET/MRI, particularly to justify the added expense and complexity of PET/MRI instead of PET/CT. This review aims to summarize the current body of evidence in support of PET/MRI, as well as current challenges and gaps in knowledge, and to identify oncologic conditions likely to benefit from its clinical use. We also present case examples to illustrate specific advantages of PET/MRI. Overall, this article should familiarize the reader with the current clinical applications of PET/MRI in oncology and provide an overview of the spe2

cific scenarios in which PET/MRI may provide added value over PET/CT or MRI alone.

CURRENT CHALLENGES Technical Considerations

Before delving into the clinical evidence, it is essential to discuss briefly the technical development of PET/MRI, so as to understand some of the inherent advantages, challenges, and limitations. The earliest approach to combining PET and MRI data was through software fusion of PET or PET/CT images with separately acquired MRI. The first combined apparatuses were sequential PET/MRI systems that consisted of individual PET and MRI elements connected by a common table. The newer integrated PET/MRI systems acquire PET and MRI data simultaneously in the same bore. This latter strategy may improve scanning efficiency and reduce misregistration (5) but requires technical adaptations of the PET components; additionally, both sequential and integrated PET/MRI systems require a novel method to correct for the attenuation of PET photons (6–8). Whereas the CT component of PET/CT directly provides electron density information that can be readily used to generate attenuation-corrected PET images, the MRI signal acquired during simultaneous PET/MRI instead correlates with proton density and tissue T1/T2 properties. Current approaches to MRI-based attenuation correction (AC) include segmentation-based and atlas-based methods (6,7). Segmentation-based AC is used clinically and relies on the Dixon method to classify voxels as soft tissue (i.e., muscle and solid organs), fat, lung, or air. In contrast to the atlas-based method, which fits pre-existing averaged imaging data sets to an acquired study and is currently used mainly in the research setting, the segmentation-based method uses each patient’s own imaging data and thus can account for large tumors, postsurgical changes, anatomic variants, and other findings not readily incorporated into imaging atlases. However, segmentation-based AC has its own set of limitations. Cortical bone, which attenuates PET photons more than soft tissue,

Academic Radiology, Vol ■, No , ■■ 2015

PET/MRI: EMERGING CLINICAL APPLICATIONS IN ONCOLOGY

does not provide adequate signal to be represented in AC maps derived from the current clinically available MRI-based segmentation methods. Consequently, cortical bone is not accounted for by the standard Dixon method, resulting in lower standardized uptake values (SUVs) for tissues within or immediately adjacent to cortical bone when assessed by PET/MRI compared to PET/CT (8). Segmentation of the lung parenchyma can occasionally fail due to the relative lack of protons available to provide MR signal compared to fat or soft tissue, resulting in artifacts that propagate into the attenuationcorrected PET images and may compromise interpretation (9). Additionally, patient-positioning devices, such as the headphones routinely used in brain MRI, can also artifactually lower SUVs derived from MR-based AC methods (10). In general, compared to CT-based AC, the maximum reductions in SUV measurements derived from MR-based AC are typically on the order of 10–20%, although the diagnostic impact of this SUV underestimation on routine clinical oncologic imaging with PET/MRI appears to be relatively minor. Workflow Optimization and Protocol Design

Given the relative novelty of PET/MRI, no standardized acquisition protocols exist. This protocol variability throughout the PET/MRI literature can make comparing the results of different studies challenging. Importantly, protocol optimization and sequence selection have been extensively described in the literature (Table 1). Regardless of the acquisition details, there are basic principles that should apply to clinical protocol development: (1) PET/MRI protocols should be designed to compete with PET/CT in terms of examination duration, and (2) PET/MRI protocols should be tailored to the clinical question at hand, with the goal of creating added value beyond what PET/CT or MRI alone might otherwise provide. At our institution, initial staging studies generally include both whole-body sequences aimed at identifying distant metastases and high-resolution, anatomically focused MR sequences in the region of the primary tumor to facilitate assessment of local invasion and detection of regional metastases. Economic Considerations

In June 2013, a Centers for Medicare & Medicaid Services (CMS) decision memo (CAG-00181R4) described a new re-

imbursement policy for FDG-PET studies, regardless of concurrent MRI or CT acquisition. Although this shift effectively cleared the path for clinical implementation of PET/MRI, dedicated billing codes for PET/MRI studies have yet to be created. Moreover, it remains unclear how private insurers, although often following the lead of CMS, will respond to clinical PET/MRI. The CMS bar for demonstrating efficacy is set quite high, requiring peer-reviewed scientific evidence to document that new technology leads to changes in patient management and improved health outcomes. Thus, future research should focus on generating outcomes data to support the clinical use of PET/MRI. EMERGING CLINICAL APPLICATIONS AND EVIDENCE The first commercial PET/MRI scanners were installed at clinical centers in 2010. In this section, we review the evidence for various oncologic applications of PET/MRI while presenting clinical cases to illustrate the scenarios in which PET/MRI is proving most useful. It should be noted that much evidence supporting the combination of PET and MRI predates the use of hybrid systems, especially in the case of neuroimaging, where software fusion of data acquired on separate PET/CT and MRI scanners is relatively straightforward. For this discussion, the terms PET/CT and PET/MRI are used to refer to these hybrid modalities without regard to a specific PET tracer. When applicable, the radiopharmaceutical used is added as a prefix (e.g., FDG-PET/MRI). Whole-body Staging

Comparing PET/MRI to PET/CT Several groups have evaluated FDG-PET/MRI in the context of a general oncology population, with various neoplasms represented and analyzed in aggregate. Early clinical evaluations of integrated whole-body PET/MRI (Fig 1) demonstrated feasibility in a general oncology population, with reasonable examination times and high-quality images (17). Subsequent studies focused on establishing the noninferiority of PET/MRI compared to PET/CT. A study of 73 consecutive patients with solid tumors that underwent routine FDGPET/CT immediately followed by integrated FDG-PET/MRI

TABLE 1. Publications Addressing PET/MRI Workflow Considerations, Protocol Design, and Sequence Optimization Fowler et al. Whole-body simultaneous positron emission tomography (PET)-MR: optimization and adaptation of MRI sequences (11). Von Schulthess et al. Workflow considerations in PET/MR imaging (12). Vargas et al. Approaches for the optimization of MR protocols in clinical hybrid PET/MRI studies (13). Barbosa et al. Workflow in simultaneous PET/MRI (14). Martinez-Möller et al. Workflow and scan protocol considerations for integrated whole-body PET/MRI in oncology (15). Kalemis et al. Sequential whole-body PET/MR scanner: concept, clinical use, and optimization after two years in the clinic. The manufacturer's perspective (5). Reiner et al. Protocol requirements and diagnostic value of PET/MR imaging for liver metastasis detection (16).

3

FRAUM ET AL

Academic Radiology, Vol ■, No , ■■ 2015

Figure 1. 29-year-old woman with newly diagnosed cervical cancer presented for initial staging. Coronal T2-weighted images (T2WIs) with 2-deoxy-2-[ 18 F]fluoro-Dglucose-positron emission tomography (FDG-PET) fusion (a posterior to b) revealed marked tracer uptake by an infiltrative mass (arrow) just inferior to the uterus (arrowhead) and posterior to the urinary bladder (asterisk), compatible with the patient’s known cervical cancer. There was no evidence of pelvic nodal disease or distant metastases. This case highlights the potential of PET/magnetic resonance imaging (MRI) to serve as a whole-body imaging modality for multiple oncologic indications.

revealed no significant difference between PET/MRI and PET/CT with respect to tumor/node/metastasis (TNM) staging accuracy, using clinical and radiological follow-up as the standard of reference (18). However, this study included only nine patients with proven metastatic disease and consequently was not powered to detect anticipated differences in sensitivity between PET/MRI and PET/CT for metastases in particular organs (e.g., better detection of liver metastases by PET/MRI, better detection of pulmonary metastases by PET/CT). A larger study of 285 general oncology patients, who underwent both FDG-PET/CT and integrated FDGPET/MRI, found that FDG-PET/CT correctly diagnosed six lesions misclassified by FDG-PET/MRI but failed to detect 30 lesions correctly recognized on FDG-PET/MRI (19). The authors conclude that FDG-PET/MRI performs better than FDG-PET/CT in most anatomical regions other than the lungs and bones. Other authors have demonstrated the potential added value of PET/MRI compared to PET/CT by analyzing the benefit of certain MR sequences, such as DWI. For example, a study comparing whole-body MRI to FDG-PET/CT in 22 patients with newly diagnosed lymphoma found that DWI resulted in disease upstaging in 23% of patients (20). These results suggest that FDG-PET/MRI that includes wholebody DWI may likewise improve lymphoma staging. Despite 4

the ability of DWI to detect metastases in organs with relatively high baseline FDG (21), some studies have failed to demonstrate that DWI incorporation into whole-body FDGPET/MRI protocols improves the lesion detection rates achieved by FDG-PET alone (22). Although there have been no large-scale prospective trials, these initial findings suggest that FDG-PET/MRI and FDG-PET/CT may have comparable accuracy for general whole-body TNM staging. However, questions remain regarding quantitative differences between PET/MRI and PET/CT, especially in light of their respective approaches to AC. A study of oncology patients with various malignancies that evaluated the SUVs of normal tissues (i.e., no malignant involvement) noted comparable values on FDG-PET/CT and FDG-PET/MRI in all organs and tissues assessed except for the lung, subcutaneous fat, and the blood pool (23). Another group has even found higher maximum SUVs (SUVmax) for nonosseous malignant lesions evaluated on FDG-PET/MRI compared to FDGPET/CT (24). As mentioned previously, osseous structures present a unique challenge for MRI-based AC. Cortical bone markedly attenuates PET photons but cannot be readily represented in standard segmentation-based AC; hence, malignant bone lesions may not be accurately depicted on PET images. Inadequate bone lesion AC by PET/MRI can result in SUV underes-

Academic Radiology, Vol ■, No , ■■ 2015

PET/MRI: EMERGING CLINICAL APPLICATIONS IN ONCOLOGY

Figure 2. 69-year-old man with known metastatic papillary thyroid carcinoma presented for restaging. Sagittal computed tomography (CT) images (a) revealed a subtle lytic lesion within the T2 vertebral body (arrow head) but a normal appearance of the T5 vertebral body (arrow). Sagittal CT images with 2-deoxy-2[18F]fluoro-D-glucose-positron emission tomography (FDG-PET) fusion (b) demonstrated an FDG-avid focus in the T2 (arrowhead) vertebral body highly suspicious for metastatic disease. In contrast, a focus of more subtly increased FDG uptake in the T5 vertebral body (arrow) without a CT correlate was felt to be indeterminate, as both metastatic disease and degenerative disease could conceivably produce this appearance. Sagittal T1weighted images (T1WIs) (c) from subsequent PET/magnetic resonance imaging (MRI) showed clear evidence of marrow replacement in the T2 (arrowhead) and T5 (arrow) vertebral bodies. Corresponding foci of FDG avidity on sagittal T1WIs with FDG-PET fusion (d) strongly supported the presence of metastases at both sites. These images show an example of the improved anatomic delineation of malignant osseous disease with PET/MRI relative to PET/CT, as well as the power of PET/MRI to distinguish osseous malignancy from degenerative remodeling.

timation by an average of 10% and as high as 22% (8). Despite this possible SUV underestimation, FDG-PET/MRI has been found to produce higher visual conspicuity (P < 0.05) of metastatic bone lesions compared to FDG-PET/CT (25). In contrast, benign bone lesions appeared more conspicuous on FDGPET/CT because of their sclerotic nature. As Beiderwellen et al. suggest, FDG-PET/MRI may prove especially useful in cases of diffuse marrow infiltration, which is generally difficult to appreciate on CT images but readily detectable on MRI. This potential advantage of FDG-PET/MRI over FDGPET/CT in assessing for osseous metastases may be even greater in the settings of physiologic marrow rebound and artificial marrow stimulation, when the corresponding FDG-PET images can be challenging to interpret because of diffuse tracer uptake. FDG-PET/MRI incorporating T1-weighted turbo spin echo (TSE) images has even been found to be superior (P = 0.0001) to FDG-PET/CT (26) for anatomic delineation of malignant bone lesions (Fig 2). Overall, despite the challenges of MRI-based AC, PET/MRI is a robust modality for the delineation of osseous metastatic disease and a reliable means of assessing tracer accumulation within various other tissue types.

PET/CT. Furthermore, because the CT component of FDGPET/CT typically contributes up to 54% to 81% of the combined radiation dose, FDG-PET/MRI has the potential to achieve significant reductions in radiation exposure without reducing the quality of anatomic images (27). Members of radiation-vulnerable populations, such as children, adolescents, and young adults with potentially curable cancers, stand to derive the most benefit from PET/MRI radiation dose reduction, especially if frequent restaging with PET is needed for optimal management. Furthermore, preliminary data from our institution and from others (28,29) have shown that FDGPET/MRI improves characterization of certain incidental lesions initially detected on FDG-PET/CT but deemed indeterminate. This advantage of PET/MRI has the potential to reduce costs and risks to the patient from the invasive procedures and/or additional or imaging studies that might otherwise be ordered for further evaluation of unexpected or ambiguous findings.

Radiation Reduction FDG-PET/MRI can provide diagnostic PET images with an equivalent dose of administered PET tracer compared to

Intracranial Neoplasms Clinical PET/CT for the initial staging of malignancy often excludes the head above the level of the skull base, as contrast-

Specific Indications

5

FRAUM ET AL

Academic Radiology, Vol ■, No , ■■ 2015

Figure 3. 8-year-old girl with suspected recurrent small cell glioma presented for restaging. (a) Transaxial contrast-enhanced T1weighted images (T1WIs) showed a heterogeneous mass (arrows) centered in the right basal ganglia. Multiple small nonenhancing areas (asterisks) interspersed among numerous enhancing foci were noted. (b) Transaxial contrast-enhanced T1WIs with 6-[18F]fluoro-3,4-dihydroxyphenylalanine-positron emission tomography (FDOPA-PET) fusion revealed avid tracer uptake by the enhancing and nonenhancing portions of this mass (arrows), illustrating transport of the FDOPA tracer into areas of tumor involvement where the blood-brain barrier was still intact. (c) Transaxial T2-weighted images (T2WIs) acquired with a fluid-attenuated inversion recovery (FLAIR) sequence demonstrated abnormal T2 prolongation within the entire area of increased FDOPA uptake (arrows), further supporting tumor infiltration into these regions. Tumor invasion into the right temporal lobe (asterisks) was also suspected, although without a corresponding focus of FDOPA uptake. This case highlights one of the advantages of PET/magnetic resonance imaging (MRI) relative to MRI alone in the setting of neuro-oncology, as certain tracers can enter the central nervous system to delineate tumor involvement in anatomic regions where the blood-brain barrier remains intact.

enhanced brain MRI is generally the preferred imaging study for evaluating primary and metastatic brain tumors. FDGPET/CT is often of limited value in this setting, as the high background FDG uptake by normal brain parenchyma reduces the conspicuity of FDG-avid neoplasms. For this reason, many PET/MRI studies of intracranial neoplasms have employed non-FDG tracers such as the radiolabeled amino acids L-[11C]methionine (MET), O-(2-[18F]fluoro-ethyl)-L-tyrosine (FET), 6-[18F]fluoro-3,4-dihydroxy-phenylalanine (FDOPA), and the hypoxia-sensitive agent [18F]fluoromisonidazole (FMISO). These tracers are currently investigational agents in the United States and not currently available for routine clinical use. Because the coregistration and software-based fusion of separately acquired PET and MRI examinations (hereafter called “software-fused PET/MRI”) can be accomplished readily within the brain because of the negligible effects of cardiac and respiratory motion, evaluation of the clinical applications of PET/MRI in neuroimaging began well before sequential and integrated PET/MRI scanners reached commercial production. Within the brain, software-fused PET/MRI can provide precise anatomic localization of tracer uptake. Importantly, radiolabeled amino acids that target the system L-amino acid transporters, such as MET, FET, and FDOPA, do not rely on breakdown of the blood-brain barrier to gain entry into the central nervous system, allowing for evaluation of the entire tumor volume including nonenhancing tumor regions (Fig 3). FDOPA-PET has even been shown to identify some gliomas missed on MRI alone (30). Similarly, studies of software-fused MET-PET/MRI have demonstrated the ca6

pacity of the MET tracer to localize both enhancing and nonenhancing glioma (31). Several other software-fused PET/MRI studies of glioma patients have found low correlation between regions of greatest tumor vascularity and metabolic activity (32,33). These results suggest that PET/MRI can provide two complementary components of imaging data, which may prove beneficial for treatment planning or response assessment. Additionally, simultaneous PET/MRI reduces the number of imaging sessions that patients must undergo and allows dynamic PET data acquisition during MRI acquisition, without extending the total imaging time. Several studies suggest that PET/MRI can potentially play an important role in guiding the management of glioma patients. A prospective study of 22 patients showed that softwarefused FMISO-PET/MRI can predict response to radiation therapy, with worse overall survival and shorter time to progression among patients with larger hypoxic glioma volumes (34). Trapped intracellularly by cells with low oxygen levels, FMISO provides an imaging marker of tumoral hypoxia, an established prognostic factor for poor response to radiation therapy (35). Furthermore, software-fused FDG-PET/MRI has been used to construct a model for assessing tumor grade, with a positive predictive value of 97–100% for the diagnosis of high-grade gliomas (36). Procedural planning with software-fused FET-PET/MRI may also increase the yield of stereotactic biopsies, with tumor regions visible on both modalities having higher histopathological grades than areas visible on FET-PET or MRI only (37). Finally, a recent study suggests that FDOPA-PET, when performed at baseline and

Academic Radiology, Vol ■, No , ■■ 2015

PET/MRI: EMERGING CLINICAL APPLICATIONS IN ONCOLOGY

Figure 4. 64-year-old woman with remote history of facial nerve sparing left parotidectomy for adenoid cystic carcinoma presented with new onset of left facial nerve paralysis. Transaxial T1-weighted images (T1WIs) with 2-deoxy-2-[ 18 F]fluoro-Dglucose-positron emission tomography (FDG-PET) fusion revealed (a) a hypermetabolic mass (asterisk) involving the superficial and deep left parotid spaces, (b) increased FDG uptake in the region of the left mental foramen (arrowhead), and (d) a focus of FDG avidity at the left mandibular foramen. (c) Transaxial T1WIs demonstrated subtle enlargement of the left mandibular foramen (arrow) compared to the contralateral side (not shown). These findings were consistent with recurrent malignancy and perineural spread along the expected course of the left inferior alveolar nerve, a diagnosis that likely would have been challenging with PET/computed tomography (CT) or magnetic resonance imaging (MRI) alone. In this regard, PET/MRI can facilitate accurate T staging of head and neck carcinoma.

at 2 weeks after beginning antiangiogenic therapy with bevacizumab, can predict overall survival in patients with recurrent high-grade gliomas, with responders having 3.5 times longer median overall survival (12.1 months vs. 3.5 months; P < 0.001) than nonresponders (38). This result indicates that FDOPA-PET/MRI may be a useful imaging examination for monitoring glioma patients on antiangiogenic agents. While these findings highlight its potential prognostic value, PET/MRI may also improve clinical outcomes for brain tumor patients in certain scenarios. A study of 33 patients with cerebral glioma that underwent surgical resection with the assistance of a navigation system employing either METPET/MRI or MRI alone found the MET-PET/MRI group to have significantly longer overall survival (39). Similarly, relative to MRI only, the use of software-fused MET-PET/MRI in gamma-knife dose and volume planning may improve survival of patients with recurrent brain metastasis (40). METPET/MRI may also be an effective tool for differentiating radiation necrosis from recurrent tumor in treated brain metastases and gliomas (41). Though these survival benefits need

to be verified by larger prospective trials, PET/MRI has shown significant potential benefits in neuroimaging by facilitating the diagnosis, response prediction, and treatment of intracranial neoplasms. Head and Neck Neoplasms The head and neck region poses unique challenges for diagnostic imaging because of its anatomic complexity and associated functional processes (42). To optimize initial staging accuracy, the evaluation of newly diagnosed head and neck carcinoma (HNC) often includes both MRI and FDG-PET/CT. Consequently, FDG-PET/MRI has been the subject of considerable interest in the realm of HNC imaging, especially with respect to T staging (Fig 4). Using histopathology as the reference standard, one study of 30 patients with new diagnoses of HNC reported superior T staging from software-fused FDG-PET/MRI compared to FDG-PET/CT, with accuracies of 87% and 67%, respectively (P = 0.041) (43). Interestingly, DWI may improve HNC lesion detection rates by FDG-PET/MRI but likely does 7

FRAUM ET AL

not significantly alter overall staging assessment, compared to FDG-PET/MRI performed without DWI (44). For restaging purposes, PET/MRI may be superior to PET/CT for characterization of indeterminate FDG uptake (i.e., questionably related to recurrence), although with no statistically significant difference in diagnostic accuracy (45). While the question of whether FDG-PET/MRI truly does improve the accuracy of T-staging will require larger studies to answer, FDG-PET/MRI for HNC has also been reported to be comparable to FDG-PET/CT in lesion conspicuity and SUVmax (46,47). Thoracic Neoplasms Because of the relatively low proton density of the lungs and resultant poor MR signal, evaluation of thoracic malignancies

Academic Radiology, Vol ■, No , ■■ 2015

relies heavily on CT and FDG-PET/CT. Consequently, FDGPET/MRI may initially seem to be a suboptimal means of characterizing neoplasms of the chest. This limitation is likely to be most pronounced in the detection of subcentimeter pulmonary nodules without increased FDG uptake (48). Despite this expectation, studies of patients undergoing staging of nonsmall cell lung cancer (NSCLC) have found integrated FDG-PET/MRI to be as accurate as FDG-PET/CT with respect to TNM staging, although the numbers of enrolled research subjects were low (49,50). Moreover, the superb softtissue contrast of MRI can be helpful in assessing invasion of adjacent structures, such as the mediastinum or chest wall (Fig 5). Interestingly, a larger prospective study of 52 patients with NSCLC comparing MRI to FDG-PET/CT found MRI to be superior to FDG-PET/CT for T staging because

Figure 5. 60-year-old woman with a remote history of treated cervical cancer was referred from an outside institution for evaluation of a lung mass. (a) Transaxial contrast-enhanced T1-weighted images (T1WIs) revealed a spiculated right parahilar lung mass (asterisk) with mediastinal invasion and near-encasement of the superior vena cava (arrow). (b) Transaxial contrast-enhanced T1WIs with 2-deoxy-2-[18F]fluoroD-glucose-positron emission tomography (FDG-PET) fusion at this same level showed the lesion (asterisk) to be hypermetabolic and also identified metastatic lesions of the thoracic spine (arrowhead) and left lung hilum (arrow). (c) Transaxial contrast-enhanced T1WIs obtained at a more superior level demonstrated lateral extension of the right parahilar lung mass (asterisk) with invasion of the chest wall (arrow). (d) Transaxial contrast-enhanced T1WIs with FDG-PET fusion at this same level revealed additional sites of metastatic disease in the right (arrowhead) and left (arrow) aspects of the mediastinum. Subsequent computed tomography (CT)-guided biopsy was positive for squamous cell carcinoma. Given the imaging appearance of the right parahilar mass and the distribution of FDG-avid lymph nodes, these findings were favored to represent metastatic bronchogenic carcinoma rather than recurrence of the patient’s treated cervical cancer. These images demonstrate the utility of PET/magnetic resonance imaging (MRI) in detecting and characterizing mediastinal and chest wall invasion by intrathoracic malignancies.

8

Academic Radiology, Vol ■, No , ■■ 2015

PET/MRI: EMERGING CLINICAL APPLICATIONS IN ONCOLOGY

of its improved detection of mediastinal and chest wall invasion but inferior for N staging because of the lower conspicuity of involved lymph nodes on MRI alone (51). These results suggest that larger prospective studies of FDGPET/MRI in NSCLC, although not yet performed, might be expected to show superior T staging and equivalent N staging compared to FDG-PET/CT. Another group that evaluated 250 patients has even found that short tau inversion recovery TSE MRI sequences, which could be readily incorporated into an FDG-PET/MRI protocol for NSCLC, may have higher sensitivity and accuracy in assessing N stage than either DWI or FDG-PET/CT (52). In contrast, a study comparing sequential FDG-PET/MRI without short tau inversion recovery TSE sequences to FDG-PET/CT found no such difference in N staging (53).

FDG-PET/MRI has also been evaluated with respect to pulmonary nodule detection. Despite the inherent challenges in MRI-based AC, one study found that integrated FDG-PET/MRI was able to detect hypermetabolic lung nodules just as well as FDG-PET/CT (48). Another study of 51 patients that underwent both FDG-PET/CT and integrated FDG-PET/MRI for staging of various cancers identified 151 pulmonary nodules on PET/CT (44 FDG avid, 107 non-FDG avid); of these 151 nodules, FDG-PET/MRI successfully detected 98% of FDG-avid nodules, 35% of nonFDG-avid nodules, 88% of nodules ≥5 mm, and 30% of nodules <5 mm (54). These results demonstrate the advantages of PET/CT for the detection of small and non-FDG-avid pulmonary nodules, although PET/MRI did perform well for FDG-avid and larger nodules.

Figure 6. 70-year-old man with cecal adenocarcinoma status post right hemicolectomy presented for restaging. (a) Transaxial computed tomography (CT) images of the liver revealed small hypodense lesions (arrows) that were deemed too small to characterize. Positron emission tomography/magnetic resonance imaging (PET/MRI) was subsequently performed for further evaluation. (b) Transaxial contrastenhanced T1-weighted images (T1WIs) obtained in the hepatocellular phase of contrast demonstrated two hypoenhancing foci (arrows) near the liver dome. (c) Transaxial contrast-enhanced T1WIs with 2-deoxy-2-[18F]fluoro-D-glucose (FDG)-PET fusion showed no definite FDGavid correlate for these lesions (arrows), likely because of their small size and relatively low FDG uptake compared to normal liver. (d) However, diffusion-weighted imaging (DWI) revealed marked diffusion restriction within these lesions (arrows) compatible with hypercellular hepatic metastases. This case demonstrates how PET/MRI, when incorporating DWI of the liver, can increase the conspicuity of (and likely also the sensitivity for) hepatic metastases, relative to PET/CT.

9

FRAUM ET AL

Academic Radiology, Vol ■, No , ■■ 2015

Beyond NSCLC, FDG-PET/MRI has been evaluated in small studies of both breast and esophageal cancer. A study of 36 patients undergoing breast cancer staging found integrated FDG-PET/MRI and FDG-PET/CT to have similar detection rates for both nodal and distant metastases (55). Another study of 49 breast cancer patients demonstrated equivalent T-staging accuracy for FDG-PET/MRI and MRI only, with both outperforming FDG-PET/CT in this regard (P < 0.05). In contrast, there were no significant differences among any of the modalities for the detection of nodal metastases (56). For esophageal cancer, sequential FDGPET/MRI and endoscopic ultrasound perform similarly with respect to both T staging and N staging, although the number of patients included in this study was low (57). Thus, the utility of FDG-PET/MRI in the evaluation of thoracic malignancies may extend beyond lung cancer to other tumor types. Abdominal Neoplasms MRI is the study of choice for evaluation of most liver lesions, given its ability via dynamic contrast enhancement and DWI to distinguish a range of benign and malignant pathologies with a high level of accuracy. Because both malignant and benign liver lesions (e.g., adenomas) can be hypermetabolic and thus can mimic metastatic disease, hepatic lesions deemed indeterminate on PET/CT are often referred to MRI for further evaluation (58). As the liver is a common site of tumor spread, especially for primary gastrointestinal cancers such as colorectal carcinoma, there has been considerable interest in examining the role of PET/MRI in the diagnosis of hepatic metastases. Several studies, including a total of 162 patients with suspected liver metastases, have found greater radiologist diagnostic confidence for liver metastasis with FDGPET/MRI compared to FDG-PET/CT (16,21,59). One of these studies, which included 37 patients and 85 liver lesions, even showed higher (P = 0.002) liver metastasis detection rates for software-fused FDG-PET/MRI (85%) versus FDGPET/CT (64%) (21). Overall, PET/MRI has the potential to improve both radiologist confidence and accuracy in diagnosing hepatic metastases, the presence versus absence of which typically alters clinical management dramatically (Fig 6). PET/MRI may also prove useful for abdominal neoplasms originating outside of the liver. One study of 119 patients with a variety of pancreatic masses found that softwarefused FDG-PET/MRI improved the accuracy of lesion classification (i.e., malignant versus benign) to 97% (P = 0.005) compared to 87% for FDG-PET/CT (60). For neuroendocrine tumors (NETs), the 68Ga-labeled somatostatin analogs (e.g., DOTATOC, DOTATATE, and DOTANOC), which take advantage of somatostatin receptor expression by NETs, have superior diagnostic performance compared to [111I]octreotide, a related compound used for clinical SPECT and single-photon emission computed tomography (SPECT)/CT imaging of NETs (61,62) (Fig 7). These newer 68 Ga-labeled PET tracers allow NET imaging at 1 hour after tracer injection, in contrast to the 18–24 hours needed prior to imaging with [111I]octreotide for optimal results. These agents 10

Figure 7. 63-year-old woman with history of neuroendocrine tumor arising from the duodenum treated with surgical resection 12 years prior, followed by recurrent oligometastatic disease to the liver status post segment VIII wedge resection 9 months prior, presented for restaging. (a) Transaxial contrast-enhanced T1-weighted images (T1WIs) of the liver revealed numerous arterially enhancing foci throughout the hepatic parenchyma (arrows) and at the site of previous segment VIII wedge resection (arrowhead). (b) Transaxial T2-weighted images (T2WIs) with DOTANOC-positron emission tomography (PET) fusion demonstrated significant tracer accumulation within the liver parenchyma (arrows) and at the site of prior surgical resection (arrowhead). DOTANOC binds to somatostatin receptor subtypes 2, 3, and 5, which are expressed by many neuroendocrine tumors (NETs). This case illustrates how the molecular properties of tumors can be exploited by radiotracers to ensure that an abnormality identified on an anatomic image in fact represents the tumor of interest rather than a benign neoplasm or a second unrelated malignancy. These images, not previously published, are courtesy of Matthias Eiber, MD, and Martin Henninger, MD, of Technische Universität München, Munich, Germany.

have also been evaluated in the context of both PET/MRI and PET/CT. One study of 24 patients reported similar conspicuity of NETs on integrated DOTATOC-PET/MRI and DOTATOC-PET/CT (63). Another study of eight patients suggested that integrated DOTATOC-PET/MRI may be superior to DOTATOC-PET/CT for detecting abdominal lesions but possibly inferior for detecting pulmonary lesions (64). Interestingly, a study of 10 patients with NETs evaluated by DOTATOC-PET/MRI and DOTATOC-PET/CT found isolated hepatobiliary phase images (obtained with gadoxetate disodium) to be significantly more sensitive (P < 0.001) for hepatic metastatic disease than the

Academic Radiology, Vol ■, No , ■■ 2015

PET/MRI: EMERGING CLINICAL APPLICATIONS IN ONCOLOGY

Figure 8. 72-year-old woman with a new diagnosis of vaginal small cell carcinoma presented for initial staging. (a) Transaxially-acquired nonisotropic T2-weighted images (T2WIs) showed a tumor (asterisk) centered along the anterior aspect of the vaginal canal (v), with extension anteriorly toward the urinary bladder (bl) and posteriorly toward the rectum (r). (b) Sagittal reformat of these nonisotropic T2WIs resulted in a significant degradation of the image quality. Assessment of the relationship between the tumor (asterisk) and the adjacent structures was difficult. Transaxially-acquired isotropic T2WIs (c) with 2-deoxy-2-[18F]fluoro-D-glucose-positron emission tomography (FDG-PET) fusion (e) showed invasion of the tumor (asterisk) into the posterior urinary bladder wall (arrow) and the anterior rectal wall (arrowhead). Sagittal reformat of these isotropic T2WIs (d) with FDG-PET fusion (f) much more clearly depicted the full craniocaudal extent of tumor (asterisk) invasion into the posterior urinary bladder wall (arrow). This case demonstrates the advantages of isotropic magnetic resonance (MR) sequences for the accurate and confident local staging of primary tumors via PET/MRI.

11

FRAUM ET AL

DOTATOC-PET images alone, likely because of high background tracer uptake or alterations in somatostatin receptor expression by the metastatic lesions (65). Furthermore, PET/MRI has been shown to be feasible for staging of paragangliomas (66). Overall, the potential applications of PET/MRI in abdominal imaging are manifold and warrant further investigation with larger prospective trials. Pelvic Neoplasms Given the anatomic complexity of the female pelvis, PET/MRI has also been examined in the context of gynecologic malignancies, including cervical cancer, ovarian cancer, and endometrial cancer. In clinical practice, MRI is often obtained as a complement to PET/CT for the initial evaluation of these tumor types, especially in light of stage-dependent treatment variations. At our institution, cervical cancer has even emerged as the most common indication for clinical FDGPET/MRI. Because of its superb soft-tissue contrast, PET/MRI

Academic Radiology, Vol ■, No , ■■ 2015

may improve the assessment of primary tumor involvement of adjacent structures, especially when sequences employing isotropic voxels with the capacity for high-resolution multiplanar reformats are used (Fig 8). Using histopathological correlation and follow-up imaging as the standards of reference, two retrospective studies of 35 cervical and 30 endometrial cancer patients found softwarefused FDG-PET/MRI to have significantly better (P = 0.0077 for cervical, P = 0.041 for endometrial) T-staging accuracy compared to FDG-PET/CT, although there were no differences in N-staging accuracy (67,68). However, the superb softtissue contrast of MRI may help to distinguish nodal involvement from physiologic metabolic activity within normal structures (Fig 9). A similar study of 26 patients with various gynecologic malignancies found T-staging accuracy by FDGPET/MRI to be superior (P < 0.001) to FDG-PET/CT (69). Beyond the time of initial diagnosis, a study of uterine, cervical, and ovarian cancer patients undergoing imaging restaging

Figure 9. 46-year old woman with newly diagnosed squamous cell carcinoma of the cervix presented for initial staging. Transaxial computed tomography (CT) images (a) with 2-deoxy-2-[18F]fluoro-D-glucose-positron emission tomography (FDG-PET) fusion (c) demonstrated a hypermetabolic soft-tissue nodule (arrow) in the region of the right external iliac artery and similar focus of less-intense uptake near the left external iliac artery (arrowhead). Differential considerations included nodal metastases, left ovarian metastases, and physiological ovarian uptake. PET/magnetic resonance imaging (MRI) was subsequently performed for further evaluation. Transaxial high-resolution T2-weighted images (T2WIs) (b) with FDG-PET fusion (d) revealed ovoid structures with internal T2-hyperintense cystic spaces compatible with ovaries. There was significant FDG uptake on the right (arrow) and trace FDG uptake on the (left), similar to the PET findings from the PET/CT examination. Consequently, nodal metastases were excluded from the differential. As with the prior case, the superior soft-tissue contrast of PET/MRI relative to PET/CT also promotes accurate N staging of gynecologic malignancies.

12

Academic Radiology, Vol ■, No , ■■ 2015

PET/MRI: EMERGING CLINICAL APPLICATIONS IN ONCOLOGY

Figure 10. 64-year-old man with an elevated prostate-specific antigen (PSA) level of 7.8 ng/mL presented for imaging evaluation of the prostate gland. Transaxial T2-weighted images (T2WIs) (a) with prostate-specific membrane antigen-positron emission tomography (PSMA-PET) fusion (b) revealed a well-circumscribed hypointense lesion (asterisk) with marked PSMA expression. This lesion was located within the right peripheral zone near the apex of the gland. (c) Apparent diffusion coefficient (ADC) map showed a corresponding dark focus (asterisk) indicative of restricted diffusion from hypercellular tumor. Relative to magnetic resonance imaging (MRI) only, PET/MRI employing prostate-specific tracers can markedly increase the conspicuity of prostate carcinoma, both within the prostate gland (as this case demonstrates) and elsewhere within the body. These images, not previously published, are courtesy of Matthias Eiber, MD, and Martin Henninger, MD, of Technische Universität München, Munich, Germany.

found software-fused FDG-PET/MRI to have a higher sensitivity (P = 0.041) for recurrent malignancy than FDGPET/CT (70). Furthermore, integrated FDG-PET/MRI has been found to increase overall radiologist confidence in diagnosing recurrent gynecologic malignancy, compared to PET/CT or MRI alone (71,72). There has also been special interest in the correlation between the tumor cellularity (as assessed by apparent diffusion coefficients) and metabolic activity (as assessed by SUVs) of primary gynecologic malignancies (73). Although the incorporation of DWI into PET/MRI protocols for pelvic neoplasms might increase radiologist diagnostic confidence, it may not improve the accuracy of malignant lesion detection (74). Like the gynecologic malignancies, prostate cancer imaging is another realm of extensive PET/MRI research efforts. In part, due to its convenience as an in-office procedure, transrectal ultrasonography has conventionally been the imaging modality used to guide biopsies of the prostate gland in evaluating for suspected malignancy. However, because transrectal ultrasonographyguided biopsies have undesirably high false-negative rates of up to 40–45%, there has been a recent push to utilize advanced imaging modalities for more reliable localization of abnormal prostatic tissue (75). Incorporating T2-weighted images, DWI, and dynamic contrast-enhanced T1-weighted images into a single MRI-based model for prostate cancer detection, prostate imaging and reporting data system (PIRADS) has provided a systematic approach for the identification of abnormal prostatic tissue on MRI (76). A natural extension of these efforts has been the addition of metabolic and molecular information from PET as a means of further improving prostate cancer diagnosis (Fig 10). Because FDG has relatively low sensitivity for detecting prostate cancer before the disease has become advanced, nonFDG-PET tracers including [11C]choline (CHO), the amino

acid anti-1-amino-3[18F]fluorocyclobutyl-1-carboxylic acid (FACBC), and small molecule prostate-specific membrane antigen ligands, have been a major focus of PET imaging in prostate cancer (77). A study of 17 patients with 51 tumor nodules suggested that software-fused CHO-PET/MRI can both differentiate Gleason ≥3 + 4 disease from Gleason ≤3 + 3 disease and improve lesion conspicuity (P < 0.01) relative to software-fused CHO-PET/CT (78). Similarly, studies of simultaneous [18F]fluorocholine-PET/MRI have shown strong correlations between various metabolic-volumetric parameters and pathology/laboratory data such as serum prostatespecific antigen, tumor volume, and Gleason score (79). Several studies of patients with newly diagnosed prostate cancer have also shown software-fused PET/MRI to improve the anatomic delineation of disease within the prostate, compared to PET/CT, PET, or MRI alone (80,81). Finally, in patients with treated prostate cancer but with rising levels of prostatespecific antigen, PET/MRI may offer higher accuracy than PET/CT or MRI alone for the imaging diagnosis of recurrence (82,83). Overall, there is great potential for PET/MRI to become an important technology for the diagnosis and staging of prostate cancer. Finally, there are few published studies on PET/MRI in the setting of colorectal and anal cancer. A pilot study of 12 patients with colorectal cancer found that FDG-PET/MRI, compared to FDG-PET/CT, provides superior T staging for some patients, though this conclusion was limited by the small sample sizes involved (84). Furthermore, experience at our institution suggests that integrated PET/MRI is useful in assessing the depth of rectal cancer invasion (Fig 11a,b), the status of locoregional lymph nodes (Fig 11c,d), and the involvement of individual anal sphincter muscles by anal carcinoma. As discussed previously, the superiority of MRI over CT for the detection and characterization of liver lesions is an additional 13

FRAUM ET AL

Academic Radiology, Vol ■, No , ■■ 2015

Figure 11. 50-year-old woman with newly diagnosed rectal adenocarcinoma presented for initial staging. Transaxial isotropic T2weighted images (T2WIs) (a) with 2-deoxy-2-[18F]fluoro-D-glucose-positron emission tomography (FDG-PET) fusion (b) showed effacement of the fat plane (long arrow) between the rectum (r) and the vagina (v) by a large, hypermetabolic tumor arising from the rectum. The fat plane between the urinary bladder (bl) and the vagina was preserved, indicating bladder sparing. Coronal reformats of these isotropic T2WIs (c) with FDG-PET fusion (d) revealed a 10 × 8 mm right internal iliac lymph node (arrowhead) that failed to meet size criteria for lymphadenopathy but displayed FDG avidity highly suspicious for nodal metastatic disease. These images demonstrate the ability of PET/magnetic resonance imaging (MRI) to detect tumor spread to lymph nodes that might not appear suspicious on MRI alone because of their normal size and/or morphology, potentially resulting in clinically significant upstaging.

advantage for whole-body staging with PET/MRI in colorectal cancer patients who are at risk for hepatic metastases. FUTURE DIRECTIONS As mentioned previously, further research is needed to provide the rigorous evidence necessary to establish PET/MRI as a routine clinical examination. Ideally, such a study would entail patients undergoing both PET/MRI and PET/CT in a randomized order, using two separate FDG administrations (i.e., one for each examination). Instead, many of the PET/MRI studies conducted thus far have employed a design in which patients undergo PET/CT followed by PET/MRI, using a single FDG administration. This model avoids exposing patients to two separate radiotracer doses but produces systematic differences in tracer 14

uptake time that confound the interpretation of results. A potential compromise might be to randomize the order in which PET/CT and PET/MRI are acquired on a patient-by-patient basis, still using a single FDG administration. This strategy would eliminate uptake time biases while still allowing for a head-tohead comparison of PET/CT and PET/MRI, without resulting in additional radiation exposure or patient inconvenience. Beyond simply comparing PET/MRI to PET/CT, a more apt analysis would involve the comparison of PET/MRI to the combination of PET/CT and MRI. Such studies would be appropriate for malignancies that typically require both PET/CT and MRI for complete staging, such as cervical cancer or colorectal cancer. Demonstration of PET/MRI noninferiority would be a reasonable goal, although there is reason to believe that PET/MRI might prove superior to the

Academic Radiology, Vol ■, No , ■■ 2015

PET/MRI: EMERGING CLINICAL APPLICATIONS IN ONCOLOGY

combination of PET/CT and MRI in terms of staging accuracy, economic efficiency, total scanner time, and patient convenience. While PET/MRI will continue to be a useful research tool, our article conveys the potential of PET/MRI to become the new standard-of-care for the diagnosing, staging, and monitoring of many different oncologic conditions (Table 2). CONCLUSION Overall, there are many potential scenarios in which PET/MRI can provide added value compared to PET/CT or MRI alone (Table 3). However, it should be noted that PET/CT still

confers certain diagnostic advantages over PET/MRI (Table 3), especially in the context of osseous or pulmonary lesions. Because of current price differences between PET/MRI and PET/CT scanners, the long-term economic viability of clinical PET/MRI will depend on scanning efficiency, perceived clinical utility, and reimbursement. Our institutional experience and a growing body of literature demonstrate the advantages of PET/MRI in numerous clinical scenarios. The advantages of PET/MRI include, but are not limited to, the definitive characterization of certain incidental lesions, the reduction of radiation doses to vulnerable populations, the improvement in accuracy of local staging, and the value that DWI adds to PET in organs with high

TABLE 2. Selected Studies Demonstrating Advantages of PET/MRI Over PET/CT Study

Primary Malignancy

Findings (PET/MRI Relative to PET/CT)

P Value

Schaarschmidt et al. (28) Catalano et al. (29) Beiderwellen et al. (25) Eiber et al. (26) Kanda et al. (43) Grueneisen et al. (56) Donati et al. (21) Nagamachi et al. (60) Queiroz et al. (69) Kitajima et al. (70) Park et al. (78)

Various Various Various Various Head and neck SCC Breast Various Pancreatic Gynecologic Gynecologic Prostate

Fewer indeterminate incidental lesions More clinically significant findings Higher conspicuity of OMs Better anatomic delineation of OMs Superior T staging accuracy Superior T staging accuracy Greater sensitivity for liver metastases Better benign vs. malignant differentiation Superior T staging accuracy Higher sensitivity for local recurrence Better identification of high-grade tumors

<0.001 <0.001 <0.05 0.0001 0.041 <0.05 0.002 0.005 <0.001 0.041 <0.01

CT, computed tomography; MRI, magnetic resonance imaging; OMs, osseous metastases; PET, positron emission tomography; SCC, squamous cell carcinoma.

TABLE 3. Comparison of PET/MRI, PET/CT, and MRI in Oncologic Imaging Potential advantages of PET/MRI over PET/CT • Better characterization of certain incidental lesions (e.g., mixed sold/cystic neoplasms, hemorrhagic or proteinaceous cysts, small cysts in solid organs, liver lesions) (28,29) • Reduction in radiation dose to vulnerable populations (e.g., children, pregnant women, young adults) of 54% to 81% (27) • Improved accuracy of local staging due to superior soft-tissue contrast (43,56,67,68) • Identification of metastatic disease in organs with high background FDG uptake (e.g., liver, brain, heart) via diffusion-weighted imaging (which detects hypercellular tumor) and dynamic contrast-enhanced MR sequences (21) • Robust MRI-based motion correction techniques, including respiratory navigation and deformable registration algorithms (85) Potential advantages of PET/MRI over MRI • Ability to assess physiologic parameters such as glucose metabolism or tumor hypoxia, which can facilitate treatment planning (34) • Identification of malignant involvement of lymph nodes that are otherwise normal in size and morphology (Fig 10) • Better anatomic delineation of gliomas because of uptake of tracers targeting the system L-amino acid transporters by nonenhancing regions of tumor involvement (30,31) Potential advantages of PET/CT over PET/MRI • Lower scanner costs and generally shorter acquisition times (11,17) • Not contraindicated in patients with pacemakers, aneurysm coils, etc. • Superior anatomic evaluation of the lung parenchyma, including greater sensitivity for pulmonary nodules that are too small to resolve by PET or MRI (54) • Less susceptible to attenuation correction artifacts, such as those arising from tissue classification errors in MRI-based segmentation algorithms (9) • Better conspicuity of benign bone lesions (25) CT, computed tomography; FDG, 2-deoxy-2-[ 18 F]fluoro-D-glucose; MRI, magnetic resonance imaging; PET, positron emission tomography.

15

Academic Radiology, Vol ■, No , ■■ 2015

FRAUM ET AL

background tracer uptake. Moreover, as new PET radiopharmaceuticals and MRI sequences become available clinically, the potential uses of PET/MRI will grow, both within and beyond the realm of oncology. Further research efforts are warranted to identify new clinical applications for PET/MRI and refine its existing roles. REFERENCES 1. Kostakoglu L, Agress H, Goldsmith SJ. Clinical role of FDG PET in evaluation of cancer patients. Radiographics 2003; 23:315–340. 2. Fletcher JW, Djulbegovic B, Soares HP, et al. Recommendations on the use of 18F-FDG PET in oncology. J Nucl Med 2008; 49:480–508. 3. Ben-Haim S, Ell P. 18F-FDG PET and PET/CT in the evaluation of cancer treatment response. J Nucl Med 2009; 50:88–99. 4. Treglia G, Sadeghi R, Del Sole A, et al. Diagnostic performance of PET/CT with tracers other than F-18-FDG in oncology: an evidence-based review. Clin Transl Oncol 2014; 16:770–775. 5. Kalemis A, Delattre BMA, Heinzer S. Sequential whole-body PET/MR scanner: concept, clinical use, and optimisation after two years in the clinic. The manufacturer’s perspective. MAGMA 2013; 26:5–23. 6. Hofmann M, Bezrukov I, Mantlik F, et al. MRI-based attenuation correction for whole-body PET/MRI: quantitative evaluation of segmentationand atlas-based methods. J Nucl Med 2011; 52:1392–1399. 7. Hofmann M, Steinke F, Scheel V, et al. MRI-based attenuation correction for PET/MRI: a novel approach combining pattern recognition and atlas registration. J Nucl Med 2008; 49:1875–1883. 8. Aznar MC, Sersar R, Saabye J, et al. Whole-body PET/MRI: the effect of bone attenuation during MR-based attenuation correction in oncology imaging. Eur J Radiol 2014; 83:1177–1183. 9. Keller SH, Holm S, Hansen AE, et al. Image artifacts from MR-based attenuation correction in clinical, whole-body PET/MRI. MAGMA 2013; 26:173–181. 10. Ferguson A, McConathy J, Su Y, et al. Attenuation effects of MR headphones during brain PET/MR studies. J Nucl Med Technol 2014; 42:93– 100. 11. Fowler KJ, McConathy J, Narra VR. Whole-body simultaneous positron emission tomography (PET)-MR: optimization and adaptation of MRI sequences. J Magn Reson Imaging 2014; 39:259–268. 12. Von Schulthess GK, Veit-Haibach P. Workflow considerations in PET/MR imaging. J Nucl Med 2014; 55:19S–24S. 13. Vargas M-I, Becker M, Garibotto V, et al. Approaches for the optimization of MR protocols in clinical hybrid PET/MRI studies. MAGMA 2013; 26:57–69. 14. Barbosa FG, von Schulthess G, Veit-Haibach P. Workflow in simultaneous PET/MRI. Semin Nucl Med 2015; 45:332–344. 15. Martinez-Möller A, Eiber M, Nekolla SG, et al. Workflow and scan protocol considerations for integrated whole-body PET/MRI in oncology. J Nucl Med 2012; 53:1415–1426. 16. Reiner CS, Stolzmann P, Husmann L, et al. Protocol requirements and diagnostic value of PET/MR imaging for liver metastasis detection. Eur J Nucl Med Mol Imaging 2014; 41:649–658. 17. Drzezga A, Souvatzoglou M, Eiber M, et al. First clinical experience with integrated whole-body PET/MR: comparison to PET/CT in patients with oncologic diagnoses. J Nucl Med 2012; 53:845–855. 18. Heusch P, Nensa F, Schaarschmidt B, et al. Diagnostic accuracy of wholebody PET/MRI and whole-body PET/CT for TNM staging in oncology. Eur J Nucl Med Mol Imaging 2015; 42:42–48. 19. Tian J, Fu L, Yin D, et al. Does the novel integrated PET/MRI offer the same diagnostic performance as PET/CT for oncological indications? PLoS ONE 2014; 9:e90844. 20. Van Ufford HMEQ, Kwee TC, Beek FJ, et al. Newly diagnosed lymphoma: initial results with whole-body T1-weighted, STIR, and diffusionweighted MRI compared with 18F-FDG PET/CT. AJR Am J Roentgenol 2011; 196:662–669. 21. Donati OF, Hany TF, Reiner CS, et al. Value of retrospective fusion of PET and MR images in detection of hepatic metastases: comparison with 18F-FDG PET/CT and Gd-EOB-DTPA-enhanced MRI. J Nucl Med 2010; 51:692–699. 22. Buchbender C, Hartung-Knemeyer V, Beiderwellen K, et al. Diffusionweighted imaging as part of hybrid PET/MRI protocols for whole-body

16

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35. 36.

37.

38.

39.

40.

41.

42. 43.

44.

cancer staging: does it benefit lesion detection? Eur J Radiol 2013; 82:877– 882. Heusch P, Buchbender C, Beiderwellen K, et al. Standardized uptake values for [18F] FDG in normal organ tissues: comparison of wholebody PET/CT and PET/MRI. Eur J Radiol 2013; 82:870–876. Iagaru A, Mittra E, Minamimoto R. Simultaneous whole-body time-of-flight 18F-FDG PET/MRI: a pilot study comparing SUVmax with PET/CT and assessment of MR image quality. Clin Nucl Med 2015; 40:1– 8. Beiderwellen K, Huebner M, Heusch P, et al. Whole-body [18F]FDG PET/MRI vs. PET/CT in the assessment of bone lesions in oncological patients: initial results. Eur Radiol 2014; 24:2023–2030. Eiber M, Takei T, Souvatzoglou M, et al. Performance of whole-body integrated 18F-FDG PET/MR in comparison to PET/CT for evaluation of malignant bone lesions. J Nucl Med 2014; 55:191–197. Huang B, Law MW-M, Khong P-L. Whole-body PET/CT scanning: estimation of radiation dose and cancer risk. Radiology 2009; 251:166– 174. Schaarschmidt BM, Grueneisen J, Heusch P, et al. Does 18F-FDG PET/MRI reduce the number of indeterminate abdominal incidentalomas compared with 18F-FDG PET/CT? Nucl Med Commun 2015; 36:588–595. Catalano OA, Rosen BR, Sahani DV, et al. Clinical impact of PET/MR imaging in patients with cancer undergoing same-day PET/CT: initial experience in 134 patients—a hypothesis-generating exploratory study. Radiology 2013; 269:857–869. Ledezma CJ, Chen W, Sai V, et al. 18F-FDOPA PET/MRI fusion in patients with primary/recurrent gliomas: initial experience. Eur J Radiol 2009; 71:242–248. Navarria P, Reggiori G, Pessina F, et al. Investigation on the role of integrated PET/MRI for target volume definition and radiotherapy planning in patients with high grade glioma. Radiother Oncol 2014; 112:425– 429. Berntsson SG, Falk A, Savitcheva I, et al. Perfusion and diffusion MRI combined with 11C-methionine PET in the preoperative evaluation of suspected adult low-grade gliomas. J Neurooncol 2013; 114:241–249. Filss CP, Galldiks N, Stoffels G, et al. Comparison of 18F-FET PET and perfusion-weighted MR imaging: a PET/MR imaging hybrid study in patients with brain tumors. J Nucl Med 2014; 55:540–545. Spence AM, Muzi M, Swanson KR, et al. Regional hypoxia in glioblastoma multiforme quantified with [18F]fluoromisonidazole positron emission tomography before radiotherapy: correlation with time to progression and survival. Clin Cancer Res 2008; 14:2623–2630. Vaupel P, Mayer A. Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metastasis Rev 2007; 26:225–239. Yoon JH, Kim J, Kang WJ, et al. Grading of cerebral glioma with multiparametric MR imaging and 18F-FDG-PET: concordance and accuracy. Eur Radiol 2014; 24:380–389. Gempt J, Soehngen E, Förster S, et al. Multimodal imaging in cerebral gliomas and its neuropathological correlation. Eur J Radiol 2014; 83:829– 834. Schwarzenberg J, Czernin J, Cloughesy TF, et al. Treatment response evaluation using 18F-FDOPA PET in patients with recurrent malignant glioma on bevacizumab therapy. Clin Cancer Res 2014; 20:3550– 3559. Tanaka Y, Nariai T, Momose T, et al. Glioma surgery using a multimodal navigation system with integrated metabolic images. J Neurosurg 2009; 110:163–172. Momose T, Nariai T, Kawabe T, et al. Clinical benefit of 11 C methionine PET imaging as a planning modality for radiosurgery of previously irradiated recurrent brain metastases. Clin Nucl Med 2014; 39:939– 943. Terakawa Y, Tsuyuguchi N, Iwai Y, et al. Diagnostic accuracy of 11Cmethionine PET for differentiation of recurrent brain tumors from radiation necrosis after radiotherapy. J Nucl Med 2008; 49:694–699. Becker M, Zaidi H. Imaging in head and neck squamous cell carcinoma: the potential role of PET/MRI. Br J Radiol 2014; 87:20130677. Kanda T, Kitajima K, Suenaga Y, et al. Value of retrospective image fusion of 18F-FDG PET and MRI for preoperative staging of head and neck cancer: comparison with PET/CT and contrast-enhanced neck MRI. Eur J Radiol 2013; 82:2005–2010. Queiroz MA, Hüllner M, Kuhn F, et al. Use of diffusion-weighted imaging (DWI) in PET/MRI for head and neck cancer evaluation. Eur J Nucl Med Mol Imaging 2014; 41:2212–2221.

Academic Radiology, Vol ■, No , ■■ 2015

PET/MRI: EMERGING CLINICAL APPLICATIONS IN ONCOLOGY

45. Queiroz MA, Hüllner M, Kuhn F, et al. PET/MRI and PET/CT in followup of head and neck cancer patients. Eur J Nucl Med Mol Imaging 2014; 41:1066–1075. 46. Partovi S, Kohan A, Vercher-Conejero JL, et al. Qualitative and quantitative performance of 18F-FDG-PET/MRI versus 18F-FDG-PET/CT in patients with head and neck cancer. AJNR Am J Neuroradiol 2014; 35:1970– 1975. 47. Varoquaux A, Rager O, Poncet A, et al. Detection and quantification of focal uptake in head and neck tumours: (18)F-FDG PET/MR versus PET/CT. Eur J Nucl Med Mol Imaging 2014; 41:462–475. 48. Rauscher I, Eiber M, Fürst S, et al. PET/MR imaging in the detection and characterization of pulmonary lesions: technical and diagnostic evaluation in comparison to PET/CT. J Nucl Med 2014; 55:724–729. 49. Schwenzer N, Schraml C, Müller M. Pulmonary lesion assessment: comparison of whole-body hybrid MR/PET and PET/CT imaging—pilot study. Radiology 2012; 264:551–558. 50. Heusch P, Buchbender C, Köhler J, et al. Thoracic staging in lung cancer: prospective comparison of 18F-FDG PET/MR imaging and 18F-FDG PET/CT. J Nucl Med 2014; 55:373–378. 51. Plathow C, Aschoff P, Lichy MP, et al. Positron emission tomography/computed tomography and whole-body magnetic resonance imaging in staging of advanced nonsmall cell lung cancer—initial results. Invest Radiol 2008; 43:290–297. 52. Ohno Y, Koyama H, Yoshikawa T, et al. N stage disease in patients with non-small cell lung cancer: efficacy of quantitative and qualitative assessment with STIR turbo spin-echo imaging, and fluorodeoxyglucose PET/CT. Radiology 2011; 261:605–615. 53. Kohan A, Kolthammer J, Vercher-Conejero J, et al. N staging of lung cancer patients with PET/MRI using a three-segment model attenuation correction algorithm: initial experience. Eur Radiol 2013; 23:3161– 3169. 54. Lee KH, Park CM, Lee SM, et al. Pulmonary nodule detection in patients with a primary malignancy using hybrid PET/MRI: is there value in adding contrast-enhanced MR imaging? PLoS ONE 2015; 10:e0129660. 55. Pace L, Nicolai E, Luongo A, et al. Comparison of whole-body PET/CT and PET/MRI in breast cancer patients: lesion detection and quantitation of 18F-deoxyglucose uptake in lesions and in normal organ tissues. Eur J Radiol 2014; 83:289–296. 56. Grueneisen J, Nagarajah J, Buchbender C, et al. Positron emission tomography/magnetic resonance imaging for local tumor staging in patients with primary breast cancer: a comparison with positron emission tomography/computed tomography and magnetic resonance imaging. Invest Radiol 2015; 50:505–513. 57. Lee G, I H, Kim S-J, et al. Clinical implication of PET/MR imaging in preoperative esophageal cancer staging: comparison with PET/CT, endoscopic ultrasonography, and CT. J Nucl Med 2014; 55:1242–1247. 58. Fosse P, Girault S, Hoareau J, et al. Unusual uptake of 18FDG by a hepatic adenoma. Clin Nucl Med 2013; 38:135–136. 59. Beiderwellen K, Gomez B, Buchbender C, et al. Depiction and characterization of liver lesions in whole body [18F]-FDG PET/MRI. Eur J Radiol 2013; 82:e669–e675. 60. Nagamachi S, Nishii R, Wakamatsu H, et al. The usefulness of (18)FFDG PET/MRI fusion image in diagnosing pancreatic tumor: comparison with (18)F-FDG PET/CT. Ann Nucl Med 2013; 27:554–563. 61. Hofmann M, Maecke H, Börner R, et al. Biokinetics and imaging with the somatostatin receptor PET radioligand (68)Ga-DOTATOC: preliminary data. Eur J Nucl Med 2001; 28:1751–1757. 62. Mayerhoefer ME, Ba-Ssalamah A, Weber M, et al. Gadoxetate-enhanced versus diffusion-weighted MRI for fused Ga-68-DOTANOC PET/MRI in patients with neuroendocrine tumours of the upper abdomen. Eur Radiol 2013; 23:1978–1985. 63. Gaertner F, Beer A, Souvatzoglou M, et al. Evaluation of feasibility and image quality of 68Ga-DOTATOC positron emission tomography/magnetic resonance in comparison with positron emission tomography/computed tomography in patients with neuroendocrine tumors. Invest Radiol 2013; 48:263–272. 64. Beiderwellen KJ, Poeppel TD, Hartung-Knemeyer V, et al. Simultaneous 68Ga-DOTATOC PET/MRI in patients with gastroenteropancreatic neuroendocrine tumors: initial results. Invest Radiol 2013; 48: 273–279.

65. Hope TA, Pampaloni MH, Nakakura E, et al. Simultaneous (68)Ga-DOTATOC PET/MRI with gadoxetate disodium in patients with neuroendocrine tumor. Abdom Imaging 2015; 40:1432–1440. 66. Blanchet E, Millo C, Martucci V. Integrated whole-body PET/MRI with 18 F-FDG, 18F-FDOPA, and 18F-FDA in paragangliomas in comparison with PET/CT. Clin Nucl Med 2013; 39:243–250. 67. Kitajima K, Suenaga Y, Ueno Y, et al. Fusion of PET and MRI for staging of uterine cervical cancer: comparison with contrast-enhanced (18)FFDG PET/CT and pelvic MRI. Clin Imaging 2014; 38:464–469. 68. Kitajima K, Suenaga Y, Ueno Y, et al. Value of fusion of PET and MRI for staging of endometrial cancer: comparison with 18F-FDG contrastenhanced PET/CT and dynamic contrast-enhanced pelvic MRI. Eur J Radiol 2013; 82:1672–1676. 69. Queiroz MA, Kubik-Huch RA, Hauser N, et al. PET/MRI and PET/CT in advanced gynaecological tumours: initial experience and comparison. Eur Radiol 2015; 25:2222–2230. 70. Kitajima K, Suenaga Y, Ueno Y, et al. Value of fusion of PET and MRI in the detection of intra-pelvic recurrence of gynecological tumor: comparison with 18F-FDG contrast-enhanced PET/CT and pelvic MRI. Ann Nucl Med 2014; 28:25–32. 71. Beiderwellen K, Grueneisen J, Ruhlmann V, et al. [(18)F]FDG PET/MRI vs. PET/CT for whole-body staging in patients with recurrent malignancies of the female pelvis: initial results. Eur J Nucl Med Mol Imaging 2015; 42:56–65. 72. Grueneisen J, Beiderwellen K, Heusch P, et al. Simultaneous positron emission tomography/magnetic resonance imaging for whole-body staging in patients with recurrent gynecological malignancies of the pelvis: a comparison to whole-body magnetic resonance imaging alone. Invest Radiol 2014; 49:808–815. 73. Grueneisen J, Beiderwellen K, Heusch P, et al. Correlation of standardized uptake value and apparent diffusion coefficient in integrated wholebody PET/MRI of primary and recurrent cervical cancer. PLoS ONE 2014; 9:e96751. 74. Grueneisen J, Schaarschmidt BM, Beiderwellen K, et al. Diagnostic value of diffusion-weighted imaging in simultaneous 18F-FDG PET/MR imaging for whole-body staging of women with pelvic malignancies. J Nucl Med 2014; 55:1930–1935. 75. Schoots IG, Roobol MJ, Nieboer D, et al. Magnetic resonance imaging– targeted biopsy may enhance the diagnostic accuracy of significant prostate cancer detection compared to standard transrectal ultrasoundguided biopsy: a systematic review and meta-analysis. Eur Urol 2015; 68:438–450. 76. Röthke M, Blondin D, Schlemmer H-P, et al. [PI-RADS classification: structured reporting for MRI of the prostate]. Rofo 2013; 185:253–261. 77. Jadvar H. Molecular imaging of prostate cancer with PET. J Nucl Med 2013; 54:1685–1688. 78. Park H, Wood D, Hussain H, et al. Introducing parametric fusion PET/MRI of primary prostate cancer. J Nucl Med 2012; 53:546–551. 79. Kim Y-I, Cheon GJ, Paeng JC, et al. Usefulness of MRI-assisted metabolic volumetric parameters provided by simultaneous (18)F-fluorocholine PET/MRI for primary prostate cancer characterization. Eur J Nucl Med Mol Imaging 2015; 42:1247–1256. 80. Hartenbach M, Hartenbach S, Bechtloff W, et al. Combined PET/MRI improves diagnostic accuracy in patients with prostate cancer: a prospective diagnostic trial. Clin Cancer Res 2014; 20:3244–3253. 81. Jambor I, Borra R, Kemppainen J, et al. Improved detection of localized prostate cancer using co-registered MRI and 11C-acetate PET/CT. Eur J Radiol 2012; 81:2966–2972. 82. Afshar-Oromieh A, Haberkorn U, Schlemmer H, et al. Comparison of PET/CT and PET/MRI hybrid systems using a 68Ga-labelled PSMA ligand for the diagnosis of recurrent prostate cancer: initial experience. Eur J Nucl Med Mol Imaging 2014; 41:887–897. 83. Piccardo A, Paparo F, Picazzo R. Value of fused 18F-choline-PET/MRI to evaluate prostate cancer relapse in patients showing biochemical recurrence after EBRT: preliminary results. BioMed Res Int 2014; 2014:103718. 84. Paspulati RM, Partovi S, Herrmann KA, et al. Comparison of hybrid FDG PET/MRI compared with PET/CT in colorectal cancer staging and restaging: a pilot study. Abdom Imaging 2015; 40:1415–1425. 85. Furst S, Grimm R, Hong I, et al. Motion correction strategies for integrated PET/MR. J Nucl Med 2015; 56:261–269.

17