Morphologic and functional imaging of malignant pleural mesothelioma

European Journal of Radiology 64 (2007) 356 366 Review Morphologic and functional imaging of malignant pleural mesothelioma Masaki Yamamuro, Victor H. Gerbaudo, Ritu R. Gill, Francine L. Jacobson, David J. Sugarbaker, Hiroto Hatabu Department of Radiology and Surgery, Brigham and Women s Hospital and Harvard Medical School, 75 Francis Street, Boston, MA 02115, United States Received 30 July 2007; received in revised form 31 July 2007; accepted 1 August 2007 Abstract Malignant pleural mesothelioma (MPM) is an aggressive tumor that arises from the pleura and frequently extends to adjacent structures. MPM cells produce and respond to many angiogenic factors, such as vascular endothelial growth factor (VEGF). VEGF expression in MPM is correlated with microvascular density, which is associated with poor survival. CT has been widely used as the primary imaging modality for the clinical evaluation of MPM. Major findings include nodular pleural thickening, unilateral pleural effusion, and tumor invasion of adjacent structures. CT tends to underestimate early chest wall invasion and peritoneal involvement and has well-known limitations in the evaluation of lymph node metastases. Perfusion CT can evaluate the microvasculature of tumors, while its disadvantages, such as high radiation exposure or side effects from iodinated contrast, limit its use in both research and clinical settings. MRI can provide additional information to CT. Because of its excellent contrast resolution, MRI is superior to CT, both in the differentiation of malignant from benign pleural disease, and in the assessment of chest wall and diaphragmatic involvement. Perfusion MRI is the most promising technique for the assessment of the tumor microvasculature. In MPM, therapeutic effects of chemotherapy can be monitored with perfusion MRI. It has been shown that FDG PET is useful for the differentiation of benign from malignant lesions, for staging and monitoring metabolic response to therapy against MPM, and that it has prognostic value. An initial report on PET/CT imaging of MPM has shown increased accuracy of overall staging, improving the assessment of tumor resectability. PET/CT seems to be superior to other imaging modalities in detecting more extensive disease involvement, and identifying unsuspected occult distant metastases. 2007 Published by Elsevier Ireland Ltd. Keywords: Malignant pleural mesothelioma; Angiogenesis; CT; MRI; PET; Perfusion; Staging Contents 1. Introduction... 357 2. MPM and angiogenesis... 357 3. MPM and imaging... 358 3.1. Computed tomography (CT)... 358 3.1.1. CT findings of MPM... 358 3.1.2. Differential diagnosis... 359 3.1.3. CT staging... 359 3.1.4. Perfusion CT... 360 3.2. Magnetic resonance imaging (MRI)... 360 3.2.1. MRI findings in MPM... 360 3.2.2. CT versus MRI... 360 Supported by the International Mesothelioma Program (IMP), Brigham and Women s Hospital. Corresponding author at: Department of Radiology and Center for Pulmonary Functional Imaging, Brigham and Women s Hospital, Harvard Medical School, 15 Francis Street, Boston, MA, United States. Tel.: +1 617 732 8353. E-mail address: hhatabu@partners.org (H. Hatabu). 0720-048X/$ see front matter 2007 Published by Elsevier Ireland Ltd. doi:10.1016/j.ejrad.2007.08.010

M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 357 3.2.3. Perfusion MRI... 361 3.3. Positron emission tomography (PET) and PET/CT... 362 3.3.1. Benign versus malignant pleural disease... 362 3.3.2. PET staging... 363 3.3.3. Predicting and monitoring therapeutic response with PET... 364 4. Future directions... 365 5. Conclusion... 365 References... 365 1. Introduction Malignant pleural mesothelioma (MPM) is an aggressive neoplasm of mesothelial cell origin that arises mainly from the pleura [1]. The major histologic subtypes are epithelial, sarcomatoid, and mixed. In addition, osteosarcomatous degeneration within MPM is considered a rare subtype. The majority of MPM cases are associated with asbestos exposure. Although MPM was once uncommon, its incidence is increasing worldwide as a result of widespread exposure to asbestos [1,2]. The tumor can invade both visceral and parietal pleura and frequently extends to adjacent structures, such as the chest wall, mediastinum, and diaphragm. Lymph node spread and/or metastases to distant organs, such as the lungs, liver, kidneys, adrenal glands, and brain, can occur. It has been shown that the overall survival in MPM is related to the extent of the primary tumor [1]. Today, imaging plays an essential role in evaluating the tumor extension of MPM. Conventional therapies, such as surgery, radiotherapy, and chemotherapy, do not necessarily improve overall survival. On the other hand, there have been a lot of advances in the understanding of the molecular biology of MPM. In particular, it has been shown that angiogenesis is directly associated with prognosis in MPM [1,3 5]. Presently, studies for the treatment of MPM with anti-angiogenesis agents are ongoing [3,6]. Imaging may play an important role in identifying the candidates for anti-angiogenesis therapy, and in documenting early response. This review discusses the current state of the science of morphologic and functional imaging of MPM in the context of diagnosis, staging and angiogenesis. 2. MPM and angiogenesis It is important to understand the molecular biological features of angiogenesis in MPM. MPM cells arise from the pleura and produce many growth factors, such as epidermal growth factor (EGF), platelet-derived growth factor (PDGF), and transforming growth factor (TGF- ) [1,3,7,8]. Genetic alterations in tumor-suppressor genes, such as p16, p14, and NF2, are common, and the activity of the antiapoptosis molecule Bcl-xL is elevated in MPM [1,9 11]. In addition, MPM cells usually express telomerase, which enable the cells to become immortalized [1,12,13]. Although the host responds to MPM cells with inflammation and immunity, growth signals coupled with the loss of tumor-suppressor genes lead to tumor cell proliferation [1,7 11]. Moreover, antiapoptosis and immortalization render MPM cells resistant to cell death [1,12,13]. The particular molecular processes can be different from patient to patient, but the overall pattern is similar. Recently, the great importance of angiogenesis in MPM has been proposed [3,6,17]. Previous studies show that the continuous formation of new blood vessels is required to supply MPM cells with nutrients and to support tumor growth by the aforementioned biological factors [1]. MPM cells produce and respond to many angiogenic factors, such as vascular endothelial growth factor (VEGF) [1,3,14,15]. In fact, MPM expresses the highest known VEGF levels of any solid tumors [2]. In addition, interleukin 8, a potent chemokine with proangiogenesis function, has been shown to be an autocrine growth factor in MPM cell lines [3,16]. VEGF expression in MPM is correlated with microvascular density, which is associated with poor survival [1,3]. Demirag et al. investigated the prognostic significance of VEGF expression in 40 MPM patients, demonstrating that VEGF expression is now considered to be an independent prognostic factor in MPM. In addition, there is a positive correlation between VEGF expression and the tumor stage (p < 0.05). Moreover, VEGF expression significantly correlates with tumor necrosis, a process known to stimulate angiogenesis (r = 0.42, p < 0.01) [17]. VEGF inhibitors have been shown to reduce MPM growth in animal models [1,3]. Studies on the treatment of MPM with anti-angiogenesis agents targeting the VEGF pathway are ongoing [2,6]. Examples of these agents are PTK787, an inhibitor of the PDGF/VEGF pathway, and Bevacizumab, a recombinant human anti-vegf monoclonal antibody [3]. Bevacizumab is now considered to be effective and generally well tolerated as first-line therapy for metastatic colon cancer, and results from clinical studies of Bevacizumab as a single agent or as part of combination regimens for breast cancer, non-small-cell lung cancer, renal cell carcinoma, and other solid malignancies have been promising [18]. Anti-angiogenic therapy inhibits the development of new tumor vessels, and destroys the existing vessels, and reduction in tumor size is not necessarily the main expected outcome. Thus, evaluation of tumor response relying solely on changes in tumor volume can lead to misinterpretation of anti-angiogenic therapeutic effects [6,19,20]. It is therefore also necessary to directly evaluate the perfusion, vascularity, vascular permeability or the biological process of angiogenesis in the tumor to assess therapeutic response.

358 M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 3. MPM and imaging 3.1. Computed tomography (CT) CT has been widely used as the primary imaging modality for the diagnosis, staging, and monitoring of therapeutic response in MPM. Typical findings include nodular pleural thickening, unilateral pleural effusion, and tumor invasion of adjacent structures. CT tends to underestimate early chest wall invasion, direct mediastinal invasion, and peritoneal involvement and has well-known limitations in the evaluation of lymph node metastases. In terms of angiogenesis, perfusion CT can evaluate the microvasculature of tumors, while its disadvantages, such as high radiation exposure, and side effects from iodinated contrast media, limit its use in both research and clinical settings. 3.1.1. CT findings of MPM MPM arises from the pleura and frequently extends to adjacent structures. Both the pleural tumor and its extension beyond the parietal pleura can be visualized by CT. Pleural thickening and/or effusions are the most common CT findings [21,22]. Several studies have illustrated the CT features of MPM. Kawashima et al. evaluated 50 MPM patients with CT and reported pleural thickening (92% of patients), thickening of the interlobular fissures (86%), pleural effusion (74%), pleural calcification (20%), contraction of the involved hemithorax (42%) and contralateral mediastinal shift (14%). They also reported disease beyond the parietal pleura, i.e., chest wall invasion (18%), pericardial effusion (6%), extension of the tumor into the contralateral hemithorax (4%), thoracic lymph nodes of 1 cm or larger in maximum diameter (58%), direct hepatic invasion (4%), and direct retroperitoneal invasion (8%) [21]. On the other hand, Sahin et al. reported CT findings in 84 patients as either unilateral pleural thickening or pleural nodules or masses (100%), pleural effusion (73%), mediastinal pleural involvement (93%), involvement of the interlobar fissures (76%), and volume contraction (i.e., reduced size of the hemithorax involved) (73%) [22]. As demonstrated in the aforementioned studies, pleural thickening and/or effusion are the most common CT findings. In the majority of MPM patients, pleural thickening is present as circumferential (rind-like) pleural involvement with multiple nodules [23] (Fig. 1). In some cases, only pleural effusion is observed by CT [1]. The combination of an unexplained pleural effusion and pleural pain should raise suspicion of MPM, even if the initial cytologic findings are negative [1]. Fig. 1. Contrast enhanced axial CT. (A C) Axial images obtained 6 months following diagnosis of MPM show circumferential pleural abnormality comprising both fluid and extensive nodular thickening, perihepatic metastases, mediastinal invasion (white arrow, B), enlarged mediastinal lymph nodes (black arrow, B) and direct tumor extension along a former chest tube tract (white arrow, C). Prior to this study, a staging cervical mediastinoscopy showed metastatic foci of MPM within enlarged 4R and level 7 nodes. The patient was not a candidate for extrapleural pneumonectomy (EPP) and declined pleurectomy/decortication, opting for chemotherapy only.

M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 359 3.1.2. Differential diagnosis CT is useful in differentiating malignant from benign disease. The presence of pleural calcifications may suggest a benign process, while that of several of the previously mentioned findings indicate malignancy; for instance, circumferential or nodular pleural thickening, or mediastinal pleural involvement [25,26]. However, MPM cannot be reliably differentiated from other pleural malignancies [27]. Diffuse pleural thickening and effusion can result from not only MPM but also from other pleural diseases, such as asbestosis, infections, and other malignant pleural tumors [24,25]. Leung et al. reviewed CT findings in 74 patients with diffuse pleural disease (11 patients with MPM, 24 with pleural metastases, and 35 with benign pleural disease), and concluded that CT is helpful in the differential diagnosis of diffuse pleural disease, particularly to differentiate malignant from benign conditions [27]. According to the literature, features that are helpful in distinguishing malignant from benign pleural disease are (a) circumferential pleural thickening, (b) nodular pleural thickening, (c) parietal pleural thickening of greater than 1 cm, and (d) mediastinal pleural involvement. Twenty-eight of 39 malignant cases (sensitivity, 72%; specificity, 83%) were identified correctly by the presence of one or more of these criteria [26]. MPM could not be reliably differentiated from pleural metastases, however [26]. On the other hand, Hierholzer et al. investigated 42 cases of pleural disease and found that circumferential pleural thickening, nodularity or irregularity of the pleural contour, mediastinal pleural involvement, and infiltration of the chest wall and/or diaphragm are most suggestive of a malignant cause, while pleural calcification in CT images is suggestive of a benign cause. However, pleural thickness of greater than 1 cm does not reveal a significant difference between malignant and benign pleural disease (p > 0.05) [27]. In general, circumferential or nodular pleural thickening is considered to be consistent with malignancy. Furthermore, infiltration of the adjacent structures may undeniably suggest malignancy. Some infectious processes, such as actinomycosis, tuberculosis and nocardiosis, can invade the chest wall, but usually do so at a single focus rather than at multiple sites, as seen in malignant disease [25]. Of note is that calcified plaques are a sign of asbestos exposure but are not a precursor to MPM [1]. However, calcification does not always indicate a benign cause, because non-calcified MPM can encase calcified plaques and mimic calcified tumors. It has been difficult to differentiate MPM from other pleural malignancies. However, Metintas et al. suggested in their recent review of 215 patients (99 patients with MPM, 39 with metastatic pleural disease and 77 with benign pleural disease) that CT findings of circumferential pleural involvement, mediastinal pleural involvement, and pleural thickness of more than 1 cm, distinguish MPM from metastatic pleural disease with relatively high sensitivity and specificity (sensitivity/specificity values of 70/85, 85/67, and 59/82, respectively). Of note is that circumferential pleural involvement was observed in 70% of MPM cases as compared to only in 15% of malignant pleural disease cases, 9% of tuberculous pleurisy cases, and 5% of asbestos-related advanced benign pleural disease cases. These findings may be important for patients in serious states or patients who do not desire to undergo invasive biopsy procedures [23]. 3.1.3. CT staging The International Mesothelioma Interest Group proposed a tumor-node-metastasis (TNM) staging system for MPM. This staging system is based on data suggesting that overall survival in MPM is related to the extent of the primary tumor (T status) and to lymph node involvement (N status), and classifies the preoperative patients as follows: patients with early disease and a potentially better prognosis (T1a and b); patients who may potentially benefit from surgery but may not necessarily be cured (T2 and T3); patients for whom surgery may have no benefit because of short survival and extensive local tumor spread (T4); patients with extensive regional node involvement (T4-N1, T4-N2, T4-N3) or distant metastases (M1), for whom surgery is not an option [28,29]. Therefore, it is important to distinguish the T1 T3 stages from the T4 disease for the prediction of resectability. On the other hand, the revised Brigham and Women s Hospital surgical staging system for MPM considers resectability, tumor histology, and nodal status, and includes four stages [30]: Stage I corresponds to disease confined within the capsule of the ipsilateral parietal pleura without adenopathy, lung, pericardium, diaphragm, or chest wall disease limited to previous biopsy sites. Stage II is the same as stage I with positive resection margins, and/or positive intrapleural lymph nodes. In stage III, there is local extension of disease into chest wall or mediastinum; into the heart; through the diaphragm or peritoneum; or with extrapleural lymph node involvement. Stage IV corresponds to the presence of distant metastatic disease. Stage III and IV tumors are considered unresectable. This staging system has been shown to stratify survival successfully. In fact, Sugarbaker and colleagues studied 183 patients with MPM treated with extrapleural pneumonectomy followed by adjuvant chemotherapy and radiotherapy, and concluded that patients with epithelial histology, margin-negative, and extrapleural node-negative status had extended survival [30]. The assessment of local invasion of the chest wall, mediastinum, and diaphragm is essential for resectability. CT criteria for resectability include (a) preservation of extrapleural fat planes; (b) absence of extrapleural soft-tissue masses; (c) normal CT attenuation values of adjacent structures, and (d) a smooth inferior diaphragmatic surface. CT criteria for unresectability include (a) invasion of extrapleural soft tissue or fat; (b) infiltration, displacement or separation of ribs by tumor; (c) bone destruction, and/or (d) tumor encasing the diaphragm. In the mediastinum, invasion of soft tissue masses without preservation of fat planes indicates unresectability. In addition, a tumor surrounding more than 50% of a mediastinal structure is suggestive of unresectability. In contrast, preservation of normal mediastinal fat without extrapleural soft tissue masses is consistent with resectable disease [31]. At the diaphragm, the most reliable findings indicative of resectability are preservation of the extrapleural fat plane between the inferior diaphragmatic surface and the adjacent abdominal organs, and a smooth inferior diaphragmatic surface. Conversely, a soft tissue mass encasing the hemidiaphragm is considered unresectable [31].

360 M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 CT does, however, have limitations, as it tends to underestimate the extent of MPM [24,25,32]. Understaging may be less important in the resectable stages (T1 T3), but assumes greater importance in the diagnosis of T4 (unresectable) disease. In addition, the evaluation of nodal involvement (N1 N3) by CT remains suboptimal because enlarged nodes alone do not prove nodal metastases [24,25,32]. Recent multi-detector row CT and multiplanar reformatting techniques may improve the visualization of lymph nodes and tumor extent, especially in regions difficult to assess with axial imaging [32]. 3.1.4. Perfusion CT Perfusion CT using iodinated contrast material can measure blood flow, blood volume and capillary permeability [6]. Perfusion CT can provide sequential acquisition of images with high temporal resolution during the passage of contrast material through tumor tissue, reflecting the microcirculation. In fact, several investigators have used perfusion CT to evaluate microvessel density of tumors [6,33]. The advantage of perfusion CT is its simple quantification and its wide availability. However, the disadvantages, such as high radiation exposure or side effects from iodinated contrast material, limit the use of perfusion CT in both research and clinical settings [34]. 3.2. Magnetic resonance imaging (MRI) MRI provides additional information to CT in some difficult cases [31]. Because of the excellent contrast resolution, MRI is superior to CT both in the differentiation of malignant from benign pleural disease and in the assessment of chest wall and diaphragmatic involvement of MPM [35] (Fig. 2). In addition, perfusion MRI is the most promising technique for evaluating angiogenesis in MPM. 3.2.1. MRI findings in MPM Compared to adjacent chest wall musculature, MPM has intermediate or slightly high signal intensity on T1-weighted images (T1-WI) and moderately high signal intensity on T2- weighted images (T2-WI) [24,25,32]. MPM signal is enhanced with the use of gadolinium-based contrast material (Gd-CM) [24,25,32]. MRI findings in MPM include diffuse pleural thickening, pleural effusion, and involvement of adjacent structures. Pleural effusion is frequently observed as focal areas of very high signal intensity on T2-WI [25]. 3.2.2. CT versus MRI MRI is superior to CT in the differentiation of malignant from benign pleural disease. High signal intensity in relation to adjacent musculature on T2-WI and/or significant contrast enhancement on T1-WI is suggestive of malignant disease. Using morphologic features in combination with the signal intensity information, MRI has a high sensitivity and specificity in the detection of pleural malignancy [26]. In addition, contrastenhanced T1 fat-suppressed sequences are the most sensitive techniques for detecting enhancement of interlobar fissures and tumor invasion of the adjacent structures [35]. Fig. 2. Gadolinium enhanced coronal T1-weighted and axial LAVA images from the previous patient 5 weeks after the initial chest radiograph reveal enhancing circumferential pleural nodules within the posterior costophrenic sulcus, extending into the major fissure without evidence of diaphragmatic or chest wall invasion. Note the preserved smooth left hemidiaphragm on the coronal image.

M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 361 Fig. 3. (A) T2-weighted coronal MR image showing direct mediastinal invasion (arrow, A); (B and C) noncontrast CT axial images showing mediastinal nodes and encasing pleural rind, but not decisive for direct mediastinal invasion. The excellent contrast resolution of MRI can improve the evaluation of tumor extension, especially to the chest wall and through the diaphragm, resulting in better prediction of overall resectability than CT [26,36] (Fig. 3). Patz et al. compared the values of the two modalities in predicting resectability. The sensitivity was high for both CT and MRI in evaluating the resectability of MPM in the diaphragm and chest wall (94% and 93% sensitivity for CT, and 100% and 100% for MRI) [31].CT and MRI provided similar information in most cases. In difficult cases, however, important complementary anatomical information was derived only from MRI. In their review of 95 patients with MPM, Heelan et al. found MRI superior to CT in revealing invasion of the diaphragm (55% accuracy for CT vs. 82% for MRI) and in showing endothoracic fascia or solitary resectable foci of chest wall invasion (46% accuracy for CT vs. 69% for MRI) [35] (Fig. 4). One of the most important factors influencing the resolution of chest MRI images is artifacts, such as susceptibility artifacts, aliasing, and motion artifacts. Thoracic motion artifacts on MRI images may be minimized with optimal cardiac gating and respiratory compensation. Subtle positive findings on MRI depend on clear resolution of the pleura and adjacent structures, and proper gating is imperative [25,35]. 3.2.3. Perfusion MRI Perfusion MRI using Gd-CM is the most promising technique for the assessment of perfusion, vascularity, and vascular permeability of tumors [34,37 40]. At present, perfusion MRI has been widely used in clinical trials with anti-angiogenic agents to assess early changes in tumor-associated vasculature [6,40]. It has been shown that anti-angiogenic effects of chemotherapy could be successfully monitored with perfusion MRI [34]. Sequential acquisition of images with a temporal resolution of approximately 1 s during intravenous administration of Gd- CM can be obtained by perfusion MRI [6,34,37 40] and the temporal passage of contrast material through tissue, including tumor tissue, reflects the microcirculation [6,34,37 40]. As a result, the pharmacokinetic analysis of perfusion MRI can provide parameters that show significant correlation with angiogenesis [6,34,37]. Gd-CM concentrates in tumors depending on tumor perfusion, vascularity and vascular permeability [6,34,37,40]. During tumor imaging, Gd-CM diffuses passively from the blood vessel into the extracellular space [6]. In the extracellular space, Gd-CM disperses freely without binding and diffuses back into the vascular space [6]. Therefore, if the extracellular space is considered as compartment 1, and the vascular space as compartment 2, a two-compartment model can be applied to the pharmacokinetic analysis of perfusion MRI [6,34]. In this analysis, the redistribution rate constant (Kep) and the elimination rate constant (Kel) are objective measurements [34,37]. Kep is the first-order rate constant for transfer of Gd-CM from compartment 2 to compartment 1 [37]. On the other hand, Kel is the first-order rate constant for elimination of Gd-CM from compartment 1 [37]. Therefore, Kel is a systemic parameter that characterizes perfusion and Kep is a tissue-specific transport parameter that characterizes both perfusion and vascular permeability of the lesion. Perfusion MRI and angiogenesis parameters can predict the therapeutic efficacy of chemotherapy in MPM. Recently, Giesel et al. evaluated the feasibility of perfusion MRI to monitor the effect of chemotherapy on MPM by comparing pharmacokinetic parameters, including Kep and Kel, to early clinical response and survival. A pharmacokinetic two-compartment model was used to analyze perfusion MRI. The study found that non-responders had a higher Kep value (3.6 min) than clinical responders (2.6 min), which correlated with shorter survival (460 days vs. 780 days, respectively). In other words, a high pre-therapeutic Kep correlated with a poor overall response to

362 M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 Fig. 4. (A) Sagittal post-contrast T1-weighted MR image through the right hemidiaphragm shows preservation of the diaphragm (white arrow). (B) T1-weighted post-contrast sagittal MR image obtained more medially shows diaphragmatic invasion and extension of tumor into the abdomen (white arrow). (C) T1 post-contrast coronal MR through the chest, posterior to the right mass demonstrating preservation of the diaphragm. (D) More anteriorly, there is obvious diaphragmatic and hepatic invasion (white arrow). chemotherapy [34]. This could be related to the fact that Kep is representative of both perfusion and vascular permeability of MPM. 3.3. Positron emission tomography (PET) and PET/CT Functional imaging with PET has facilitated the non-invasive evaluation of tumor pathophysiology, metabolism and proliferation. The most common PET tracer used in clinical practice is 18 F-Fluorodeoxyglucose (FDG), due to its favorable halflife of 18 F (109 min). The basis for the use of FDG is the increased glucose metabolism of cancer cells compared to normal tissues [41]. Lesion uptake of FDG reflects the alteration of glycolytic metabolism in various tumors, including MPM. In addition, FDG PET imaging captures alterations in tissue metabolism that generally precede morphologic change. Today, FDG PET has achieved wide availability not only in research, but in the clinical setting as well. Preliminary reports describing the role of FDG and PET in pleural malignancies are limited but very encouraging. It has been shown that FDG PET is useful for the differentiation of benign from malignant lesions, for staging and monitoring metabolic response to therapy of MPM, and that it has prognostic value [42 46]. 3.3.1. Benign versus malignant pleural disease FDG PET imaging has been shown to accurately differentiate benign pleural lesions from MPM. Bénard et al. [42] described encouraging results in 28 patients referred for the evaluation of pleural disease and suspected MPM. Using FDG PET with visual analysis, the sensitivity was 92%, with a specificity of 75%, and an accuracy of 89% for the detection of MPM. By using semi-quantitative analysis they concluded that a standardized uptake value of tracer (SUV) greater than 2.0 differentiated between benign from malignant disease, increasing the sensitivity, specificity, and overall accuracy to 91%, 100%, and 92%, respectively. Schneider et al. [43] reported the results of a retrospective analysis of 18 consecutive patients with biopsy-proven MPM who underwent FDG PET for staging. All primary lesions were FDG-avid, distinguishing between benign and malignant pleural disease. Our results have been similar [44]. In our hands, the overall sensitivity, specificity, and accu-

racy of FDG imaging for the differential diagnosis of MPM has been 97%, 80% and 94%, respectively, compared to 83%, 80%, and 82% for diagnostic CT. Thirty-four lesions were biopsied, and FDG imaging correctly identified 28 out of 29 malignant tumors, yielding negative results in 4 out of 5 benign lesions. The smallest lesion detected was 0.8 cm. The only false-positive finding was an area of diffusely low uptake in the right costophrenic sulcus of a patient treated with talc pleurodesis due to recurrent, symptomatic pleural effusions. Other reported sources of false-positive findings include benign inflammatory pleuritis; benign asbestos-related plaques; parapneumonic effusion, and tuberculous pleuritis [42,43,45]. True-negative findings have been reported in mild pleural inflammatory reactions, benign angiolipoma, benign asbestos pleural disease, chronic fibrosing pleuritis, and pleural fibroma [42,44,45]. Although most mesotheliomas are highly FDG-avid, mild or lack of tracer uptake has been reported in patients with mesothelioma of the epithelial subtype [42]. Neither the pattern nor the intensity of FDG lesion uptake is able to differentiate between subtypes of malignant pleural mesothelioma, or differentiate MPM from adenocarcinoma or from sarcoma. Therefore, invasive techniques are still warranted. M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 363 3.3.2. PET staging Accurate staging of MPM is imperative to ascertain prognosis and to elect the appropriate therapeutic algorithm that should follow. PET imaging is an accurate and useful imaging technique for the staging and preoperative evaluation of disease extent. In most instances, PET imaging correctly defines the metabolic extent of intrathoracic involvement, providing complementary information to anatomical imaging modalities for the assessment of tumor resectability [44,46]. Generally, there is good agreement between FDG images and the extent of disease determined surgically. PET images correctly predict unresectability by detecting or confirming the presence and extent of metabolically active tumor invading mediastinal organs, transdiaphragmatic spread, and distant metastatic disease [42 46]. The typical appearance is that of moderate to high 18 F-FDG uptake involving the areas of thickened pleura. Gerbaudo et al. [46] described different patterns of FDG uptake in MPM, which could be present as focal, linear, mixed or encasing distributions of the tracer [44,46] (Fig. 5). These patterns closely reflect the extent of pleural and parenchymal involvement observed in anatomical imaging and surgery. Tumors at an earlier stage tend to have focal or linear patterns of uptake, whereas the mixed and encasing patterns are indicative of more advanced disease. These investigators also reported the results of semi-quantitative analysis of serial dual-phase FDG images, which showed that radiotracer uptake increases over time in normal tissue, and in MPM [46]. In the normal lung, the rise in FDG uptake between early and late images was 6 ± 4% in 2 h. On the other hand, the rise in FDG uptake in MPM was higher in stage IV patients (97 ± 25%), when compared to stage I (13 ± 1%), stage II (34 ± 2%), and stage III patients (57 ± 3%). Bénard et al. [42] evaluated the diagnostic value of FDG PET in 28 patients with suspected MPM. The SUV in FDG PET was Fig. 5. Coronal maximum intensity projection (MIP) ninety minutes postinjection of FDG demonstrating the encasing pattern of distribution in pleural mesothelioma with probable diaphragmatic extension (arrow). significantly higher in malignant (n = 24; SUV = 4.9 ± 2.9) than in benign lesions (n = 4; SUV = 1.4 ± 0.6) (p < 0.01). Nine out of 12 hypermetabolic lymph nodes were detected on FDG PET images that appeared normal on CT scans. The study concluded that FDG PET imaging is a sensitive method to identify malignant mesothelioma and to determine the extent of the disease process. Recently, integrated PET/CT systems have been introduced, and their use has rapidly been incorporated at the forefront of cancer imaging [47 49]. PET/CT imaging combines metabolic with anatomical information in a single imaging procedure. Recent reports suggest that integrated PET/CT scanning improves the diagnostic and staging accuracy over PET or CT alone in different malignancies [50 52]. The advantages of integrated PET/CT imaging for staging and assessing resectability of MPM have recently been reported by Erasmus et al. [53]. The authors concluded that PET/CT increases the accuracy of overall staging and improves the selection of patients for extrapleural pneumonectomy. PET/CT was superior to other imaging modalities, as it detected more extensive disease involvement, and identified occult distant

364 M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 metastases not suspected after conventional imaging. The main limitation of the technique was its inaccuracy in the detection of nodal MPM metastases. N1 disease with mild or low FDG uptake could be a source of false-negative results [53,54]. Uptake in N1 nodes might be difficult to distinguish from uptake in the adjacent pleural tumor. N2 nodes affected with micrometastatic disease could yield false-negative findings as well. In addition, false-positive results could occur in granulomatous lymphadenitis. These are certainly very important limitations, considering that N2 positive nodes are always associated with poor survival, and that N2, rather than N1 disease, carries the worst prognosis [30,55]. Both false-negative and false-positive findings in extrapleural nodes contribute misleading information with regard to outcome, and optimal treatment strategy [30]. In our experience, the use of FDG PET/CT imaging has been very useful for the selection of the most metabolically active site for needle or thoracoscopic biopsy in patients with MPM. This approach tends to increase the yield of positive biopsies, especially in those cases with diffuse pleural thickening, or with distorted anatomy after treatment and suspected recurrent disease [56]. The main limitation of PET is its limited anatomical landmarks, especially in the setting of MPM [42,44,46]. Thus, it still remains to be shown if the use of PET/CT and fusion imaging could improve the assessment of local invasion into the chest wall, pericardium, and diaphragm, during the initial staging of patients with MPM. 3.3.3. Predicting and monitoring therapeutic response with PET Patients with lesions that are highly FDG-avid have a poor prognosis. Bénard et al. [57] followed up 17 MPM patients for survival analysis. An overall median SUV of 4.03 was chosen as a cutoff to establish two patient groups: those with low-suv lesions and those with high-suv lesions. The investigators found a wide variability in SUV measurements in patients with malignant disease, and demonstrated that high levels of FDG uptake in mesothelioma were associated with an unfavorable prognosis. The mean SUV of survivors was 3.2 (1.6), compared to 6.6 (2.9) for the non-survivors. The cumulative survival estimate of the high SUV group (0.17 at 12 months) was significantly shorter than that of the low SUV group (0.86 at 12 months) (p < 0.01) [57]. Recently, Flores et al. [58] reported on the prognostic value of PET imaging in 137 patients with MPM. The authors concluded that a lesion with SUV >10, mixed histology, and stages III and IV carry the worst prognosis. Fig. 6. Coronal MIP image 90 minutes post injection of FDG. (A) Pre-treatment; (B) Post-treatment, Fused axial PET CT images; (C) Pre-treatment and (D) Post-treatment images of the same patient showing encasing pattern of distribution in malignant pleural mesothelioma with probable diaphragmatic involvement. Note reduced post-treatment FDG avidity of the right pleural mass. The patient subsequently underwent successful EPP.

M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 365 The use of FDG PET for the assessment of response to treatment in MPM appears promising. Ceresoli et al. [59] studied the predictive value of FDG PET in 20 patients to assess treatment efficacy after 2 cycles of single agent pemetrexed, or pemetrexed in combination with carboplatin. The authors defined complete metabolic response as total resolution of FDG uptake within the tumor volume between the pre- and the post-therapeutic scans (Fig. 6). Partial response was defined as a reduction in tumor FDG uptake of 25% or more, whereas progressive disease corresponded to an increase in tumor SUV of 25%, or the appearance of any new FDG-avid lesions. Stable metabolic disease was classified as an increase or decrease in tumor SUV of <25%. Early metabolic response was significantly correlated to median time-to-tumor progression (TTP); 14 months for metabolic responders, compared to 7 months for non-responders (p < 0.05). In addition, patients with a metabolic response had a trend toward longer overall survival. Of note is that no correlation was found between TTP and radiologic response assessed by CT. 4. Future directions In MPM, perfusion MRI is valuable for predicting the therapeutic efficacy of chemotherapy, although, this concept requires further examination in a larger population. In addition, direct comparison between perfusion MRI parameters and angiogenesis factors, such as VEGF expression, is necessary. Furthermore, the ability of perfusion MRI to monitor the therapeutic effect of combination chemotherapy and anti-angiogenesis drugs remains to be assessed. Other perfusion parameters, such as maximum enhancement ratio, time at maximum enhancement ratio, and slope may be associated with angiogenesis [39,60]. In fact, the usefulness of dynamic MRI and of these parameters in differentiating benign from malignant solitary pulmonary nodules has already been suggested [39,60]. In MPM, these parameters should be studied in the context of angiogenesis factors and prognosis. MRI state-of-the-art technology and imaging techniques are developing rapidly, as shown by the advent of 3T MRI and parallel imaging. Further refinement of MRI techniques and equipment may improve the temporal resolution of sequential data acquisition. In addition, three-dimensional MRI (3D-MRI), with multi-channel surface coils and/or parallel imaging techniques may be used to evaluate whole lung perfusion. These state-of-the-art techniques will be validated in future studies. PET is able to study the biological processes of tumor angiogenesis in vivo. Tracers of blood flow, vascular volume, microvessel density, hypoxia and even cell proliferation and apoptosis have already been developed and analyzed [6]. Furthermore, results from recent investigations of molecular processes and targets are being used to design new compounds. For example, the anti-vegf monoclonal antibody, VG76e, has been labelled with the PET tracer 124 I, and its imaging potential was validated successfully in a mouse model [6,61]. In addition, a radiolabelled ligand of V 3 integrin receptor, which is expressed specifically in endothelial cells of tumor vessels, has been used successfully to image tumor-associated angiogenesis [6]. These tracers may represent the future of in vivo molecular imaging of angiogenesis. However, their availability is still very limited, and to the best of our knowledge, there are no studies assessing the role of these tracers in MPM. 5. Conclusion In conclusion, CT plays a primary role in the diagnosis and staging of MPM. MRI and PET can provide additional information to help overcome the limitations of CT. Complementing CT with these other imaging techniques is promising for the assessment of angiogenesis and the prognosis of MPM. References [1] Robinson BW, Lake RA. Advances in malignant mesothelioma. N Engl J Med 2005;353(15):1591 603. [2] Curran D, Sahmoud T, Therasse P, van Meerbeeck J, Postmus PE, Giaccone G. Prognostic factors in patients with pleural mesothelioma: the European Organization for Research and Treatment of Cancer experience. J Clin Oncol 1998;16(1):145 52. [3] Zucali PA, Giaccone G. Biology and management of malignant pleural mesothelioma. Eur J Cancer 2006;42(16):2706 14. [4] Edwards JG, Cox G, Andi A, et al. Angiogenesis is an independent prognostic factor in malignant mesothelioma. Br J Cancer 2001;85(6):863 8. [5] Strizzi L, Catalano A, Vianale G, et al. Vascular endothelial growth factor is an autocrine growth factor in human malignant mesothelioma. J Pathol 2001;193(4):468 75. [6] Drevs J, Schneider V. The use of vascular biomarkers and imaging studies in the early clinical development of anti-tumor agents targeting angiogenesis. J Intern Med 2006;260(6):517 29. [7] Dazzi H, Hasleton PS, Thatcher N, Wilkes S, Swindell R, Chatterjee AK. Malignant pleural mesothelioma and epidermal growth factor receptor (EGF-R): relationship of EGF-R with histology and survival using fixed paraffin embedded tissue and the F4, monoclonal antibody. Br J Cancer 1990;61:924 6. [8] Marzo AL, Fitzpatrick DR, Robinson BWS, Scott B. Antisense oligonucleotides specific for transforming growth factor beta2 inhibit the growth of malignant mesothelioma both in vitro and in vivo. Cancer Res 1997;57:3200 7. [9] Wong L, Zhou J, Anderson D, Kratzke RA. Inactivation of p16ink4a expression in malignant mesothelioma by methylation. Lung Cancer 2002;38:131 6. [10] Yang CT, You L, Yeh CC, et al. Adenovirus-mediated p14(arf) gene transfer in human mesothelioma cells. J Natl Cancer Inst 2000;92:636 41. [11] Schipper H, Papp T, Johnen G, et al. Mutational analysis of the nf2 tumor suppressor gene in three subtypes of primary human malignant mesotheliomas. Int J Oncol 2003;22:1009 17. [12] Broaddus VC, Dansen TB, Abayasiriwardana KS, et al. Bid mediates apoptotic synergy between tumor necrosis factor related apoptosis-inducing ligand (TRAIL) and DNA damage. J Biol Chem 2005;280:12486 93. [13] Dhaene K, Wauters J, Weyn B, Timmermans JP, van Marck E. Expression profile of telomerase subunits in human pleural mesothelioma. J Pathol 2000;190:80 5. [14] Ohta Y, Shridhar V, Bright RK, et al. VEGF and VEGF type C play an important role in angiogenesis and lymphangiogenesis in human malignant mesothelioma tumours. Br J Cancer 1999;81(1):54 61. [15] Kumar-Singh S, Weyler J, Martin MJ, Vermeulen PB, Van Marck E. Angiogenic cytokines in mesothelioma: a study of VEGF, FGF-1 and -2, and TGF beta expression. J Pathol 1999;189(1):72 8. [16] Galffy G, Mohammed KA, Dowling PA, Nasreen N, Ward MJ, Antony VB. Interleukin 8: an autocrine growth factor for malignant mesothelioma. Cancer Res 1999;59:367 71. [17] Demirag F, Ünsal E, Yilmaz A, Çağlar A. Prognostic significance of vascular endothelial growth factor, tumor necrosis, and mitotic activity index in malignant pleural mesothelioma. Chest 2005;128(5):3382 7.

366 M. Yamamuro et al. / European Journal of Radiology 64 (2007) 356 366 [18] Shih T, Lindley C. Bevacizumab: an angiogenesis inhibitor for the treatment of solid malignancies. Clin Ther 2006;28(11):1779 802. [19] Byrne MJ, Nowak AK. Modified RECIST criteria for assessment of response in malignant pleural mesothelioma. Ann Oncol 2004;15(2): 257 60. [20] Oxnard GR, Armato III SG, Kindler HL. Modeling of mesothelioma growth demonstrates weaknesses of current response criteria. Lung Cancer 2006;52(2):141 8. [21] Kawashima A, Libshitz HI. Malignant pleural mesothelioma: CT manifestations in 50 cases. AJR Am J Roentgenol 1990;155(5):965 9. [22] Sahin AA, Coplu L, Selcuk ZT, et al. Malignant pleural mesothelioma caused by environmental exposure to asbestos or erionite in rural Turkey: CT findings in 84 patients. AJR Am J Roentgenol 1993;161(3):533 7. [23] Metintas M, Ucgun I, Elbek O, et al. Computed tomography features in malignant pleural mesothelioma and other commonly seen pleural diseases. Eur J Radiol 2002;41(1):1 9. [24] Roach HD, Davies GJ, Attanoos R, Crane M, Adams H, Phillips S. Asbestos: when the dust settles an imaging review of asbestos-related disease. Radiographics 2002;22:S167 84. [25] Lorenzo B, Feragalli B, Sacco R, Merlino B, Storto ML. Malignant pleural disease. Eur J Radiol 2000;34(2):98 118. [26] Hierholzer J, Luo L, Bittner RC, et al. MRI and CT in the differential diagnosis of pleural disease. Chest 2000;118(3):604 9. [27] Leung AN, Muller NL, Miller RR. CT in differential diagnosis of diffuse pleural disease. AJR Am J Roentgenol 1990;154(3):487 92. [28] Rusch VW. A proposed new international TNM staging system for malignant pleural mesothelioma. From the International Mesothelioma Interest Group. Chest 1995;108(4):1122 8. [29] Patz EF, Rusch VW, Heelan R. The proposed new international TNM staging system for malignant pleural mesothelioma: application to imaging. Am J Roentgenol 1996;166(2):323 7. [30] Sugarbaker DJ, Flores RM, Jaklitsch MT, et al. Resection margins, extrapleural nodal status, and cell type determine postoperative long-term survival in trimodality therapy of malignant pleural mesothelioma: results in 183 patients. J Thorac Cardiovasc Surg 1999;117(1):54 63. [31] Patz EF, Shaffer K, Piwnica-Worms DR, et al. Malignant pleural mesothelioma: value of CT and MR imaging in predicting resectability. AJR Am J Roentgenol 1992;159(5):961 6. [32] Wang ZJ, Reddy GP, Gotway MB, et al. Malignant pleural mesothelioma: evaluation with CT, MR imaging, and PET. Radiographics 2004;24(1): 105 19. [33] Ding B, Ling HW, Chen KM, Jiang H, Zhu YB. Comparison of cerebral blood volume and permeability in preoperative grading of intracranial glioma using CT perfusion imaging. Neuroradiology 2006;48(10):773 81. [34] Giesel FL, Bischoff H, von Tengg-Kobligk H, et al. Dynamic contrastenhanced MRI of malignant pleural mesothelioma: a feasibility study of noninvasive assessment, therapeutic follow-up, and possible predictor of improved outcome. Chest 2006;129(6):1570 6. [35] Heelan RT, Rusch VW, Begg CB, Panicek DM, Caravelli JF, Eisen C. Staging of malignant pleural mesothelioma: comparison of CT and MR imaging. AJR Am J Roentgenol 1999;172(4):1039 47. [36] Knuuttila A, Kivisaari L, Kivisaari A, et al. Evaluation of pleural disease using MR and CT. With special reference to malignant pleural mesothelioma. Acta Radiol 2001;42(5):502 7. [37] Brix G, Semmler W, Port R, Schad LR, Layer G, Lorenz W. Pharmacokinetic parameters in CNS Gd-DTPA enhanced MR imaging. J Comput Assist Tomogr 1991;15(4):621 8. [38] Hatabu H, Gaa J, Kim D, Li W, Prasad PV, Edelman RR. Pulmonary perfusion: quantitative assessment with dynamic contrast-enhanced MRI using ultra-short TE and invasion recovery turbo FLASH. Magn Reson Med 1996;36:503 8. [39] Ohno Y, Hatabu H, Takenaka D, Adachi S, Kono M, Sugimura K. Solitary pulmonary nodules; potential role of dynamic MR imaging in management initial experience. Radiology 2002;224:503 11. [40] Hylton N. Dynamic contrast-enhanced magnetic resonance imaging as an imaging biomarker. J Clin Oncol 2006;24:3293 8. [41] Warburg O, Posener K, Negelein E. The metabolism of the carcinoma cell. In: Warburg O, editor. The metabolism of tumors. New York, NY: Richard R. Smith, Inc.; 1931. p. 129 69. [42] Bénard F, Sterman D, Smith RJ, Kaiser LR, Albelda SM, Alavi A. Metabolic imaging of malignant pleural mesothelioma with fluorodeoxyglucose positron emission tomography. Chest 1998;114(3): 713 22. [43] Schneider DB, Clary-Macy C, Challa S, et al. Positron emission tomography with F18-fluorodeoxyglucose in the staging and preoperative evaluation of malignant pleural mesothelioma. J Thorac Cardiovasc Surg 2000;120:128 33. [44] Gerbaudo VH, Sugarbaker DJ, Britz-Cunningham S, Di Carli MF, Mauceri C, Treves ST. Assessment of malignant pleural mesothelioma with 18 F- FDG dual-head gamma-camera coincidence imaging: comparison with histopathology. J Nucl Med 2002;43:1144 9. [45] Bury T, Paulus P, Dowlati A, Corhay JL, Rigo P, Radermecker MF. Evaluation of pleural diseases with FDG PET imaging: preliminary report. Thorax 1997;52:187 9. [46] Gerbaudo VH, Britz-Cunningham S, Sugarbaker DJ, Treves ST. Metabolic significance of the pattern, intensity and kinetics of 18F-FDG uptake in malignant pleural mesothelioma. Thorax 2003;58(12):1077 82. [47] Kinahan PE, Townsend DW, Beyer T, Sashin D. Attenuation correction for a combined 3D PET/CT scanner. Med Phys 1998;25(10):2046 53. [48] Beyer T, Townsend DW, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41(8):1369 79. [49] Townsend DW, Cherry SR. Combining anatomy and function: the path to true image fusion. Eur Radiol 2001;11(10):1968 74. [50] von Schulthess GK, Steinert HC, Hany TF. Integrated PET/CT: current applications and future directions. Radiology 2006;238(2):405 22. [51] Lardinois D, Weder W, Hany TF, et al. Staging of non-small-cell lung cancer with integrated positron-emission tomography and computed tomography. N Engl J Med 2003;348:2500 7. [52] Antoch G, Stattaus J, Nemat AT, et al. Nonsmall cell lung cancer: dualmodality PET/CT in preoperative staging. Radiology 2003;229(2):526 33. [53] Erasmus JJ, Truong MT, Smythe WR, et al. Integrated computed tomography-positron emission tomography in patients with potentially resectable malignant pleural mesothelioma: staging implications. J Thorac Cardiovasc Surg 2005;129(6):1364 70. [54] Gerbaudo VH. 18F-FDG imaging of malignant pleural mesothelioma: scientiam impendere vero. Nucl Med Commun 2003;24(6):609 14. [55] Sugarbaker DJ, Strauss GM, Lynch TJ, et al. Node status has prognostic significance in the multimodality therapy of diffuse, malignant mesothelioma. J Clin Oncol 1993;11:1172 8. [56] Gerbaudo VH, Mamede M, Trotman-Dickenson B, Bueno R, Sugarbaker DJ. PET/CT patterns of treatment failure of malignant pleural mesothelioma. J Nucl Med 2007;48:360 (abs). [57] Bénard F, Sterman D, Smith RJ, Kaiser LR, Albelda SM, Alavi A. Prognostic value of FDG PET imaging in malignant pleural mesothelioma. J Nucl Med 1999;40(8):1241 5. [58] Flores RM, Akhurst T, Gonen M, et al. Positron emission tomography predicts survival in malignant pleural mesothelioma. J Thorac Cardiovasc Surg 2006;132(4):763 8. [59] Ceresoli GL, Chiti A, Zucali PA, et al. Early response evaluation in malignant pleural mesothelioma by positron emission tomography with [18F] fluorodeoxyglucose. J Clin Oncol 2006;24(28):4587 93. [60] Kono R, Fujimoto K, Terasaki H, et al. Dynamic MRI of solitary pulmonary nodules: comparison of enhancement patterns of malignant and benign small peripheral lung lesions. Am J Roentgenol 2007;188(1): 26 36. [61] Collingridge DR, Carroll VA, Glaser M, et al. The development of [(124)I]iodinated-VG76e: a novel tracer for imaging vascular endothelial growth factor in vivo using positron emission tomography. Cancer Res 2002;62:5912 9.