Molecular Imaging and Biology

, Volume 13, Issue 5, pp 801–811

Current Evidence Base of FDG-PET/CT Imaging in the Clinical Management of Malignant Pleural Mesothelioma: Emerging Significance of Image Segmentation and Global Disease Assessment

Authors

  • Sandip Basu
    • Radiation Medicine Center (Bhabha Atomic Research Center)Tata Memorial Hospital Annexe
  • Babak Saboury
    • Division of Nuclear Medicine, Department of RadiologyHospital of the University of Pennsylvania
  • Drew A. Torigian
    • Division of Nuclear Medicine, Department of RadiologyHospital of the University of Pennsylvania
    • Division of Nuclear Medicine, Department of RadiologyHospital of the University of Pennsylvania
Review Article

DOI: 10.1007/s11307-010-0426-6

Cite this article as:
Basu, S., Saboury, B., Torigian, D.A. et al. Mol Imaging Biol (2011) 13: 801. doi:10.1007/s11307-010-0426-6

Abstract

Increasingly, integrated positron emission tomography-computed tomography (PET/CT) imaging is playing a crucial role in the assessment of patients with known or suspected malignant pleural mesothelioma (MPM). Based on the data reported in the literature, this combined modality is likely to become the instrument of choice for examining patients of MPM. The research on this subject has focused on the following five domains: (1) differentiation of MPM from other benign pleural diseases, (2) preoperative staging for the selection of appropriate candidates for surgery, (3) evaluation for therapy response and post-treatment surveillance for recurrence, (4) prognostication based upon the intensity of 2-deoxy-2-[18F]fluoro-d-glucose (FDG) uptake, and (5) planning of radiotherapy. These represent the bases for critical decision making in the management of mesothelioma, and FDG-PET/CT offers potential advantages over conventional CT imaging and thus can play a pivotal role in this regard. Optimal characterization of this potentially fatal disease with a high negative predictive value for MPM, superior capability for cancer staging initially and at the later course of disease, and ability for measuring therapeutic response and the precise determination of the target volume for radiotherapy planning represent distinct advantages of this promising molecular imaging tool. In this communication, we have explored the promising role of integrated FDG-PET/CT in the overall management of this serious malignancy. From the available data, the major role of PET-CT at present appears to be in the preoperative disease staging, response to treatment assessment, and post-treatment disease surveillance of MPM. In all these three areas, PET-CT convincingly shows better results than conventional anatomical imaging alone and thereby can aid in exploring novel therapeutic approaches. Disease prognosis and radiotherapy planning are evolving areas where this modality has demonstrated significant promise, but this has to be investigated further. The differentiating of MPM from benign pleural disease is a challenging issue; though in limited studies, it has shown promising results, single standardized uptake value (SUV) cutoff technique cannot be the optimal way for this purpose. Dual time point and delayed imaging helps further in this setting; however, more data require to be accrued in this area. We, in this review, have also discussed the feasibility of a new method of image segmentation based on an iterative thresholding algorithm, which permits definition of the boundaries of lesions based on PET images alone to provide lesional metabolically active tumor volumes, lesional partial volume corrected SUV (PVC-SUV) measurements, lesional PVC metabolic burden (PVC-MB) (calculated as the product of lesional MVP and lesional PVC-SUV), and whole body metabolic burden (WBMB) (calculated as the sum of lesional PVC-MB of all lesions). This global disease assessment, we believe, will be the way forward for assessing this malignancy with a non-invasive imaging modality.

Key Words

PETPET/CTMesotheliomaFDGPleural mesotheliomaMalignant pleural mesothelioma

PET and PET/CT Imaging in MPM: What are the Current Needs?

Malignant pleural mesothelioma (MPM) is an aggressive malignancy that is almost uniformly fatal within a year of diagnosis. The annual incidence of this malignancy in the United States is estimated to be 2,200 cases per year. There has been an increased incidence of 50% in the past decade [1, 2]. A similar rise in the incidence of mesothelioma is expected in the next few years in Europe due to the 20–30-year time lag in the development of MPM following asbestos exposure in the 1970s–1980s [3]. Presently, the standard of care for first-line therapy is a combination chemotherapy comprised of cisplatin and pemetrexed. In a study of 456 patients, 226 received this regimen and demonstrated a response rate of 41% versus 17% with cisplatin alone and a median survival of 12.1 and 9.3 months, respectively [4, 5]. Symptom-free disease progression has also been shown to be improved with implementation of chemotherapy early after diagnosis [6]. Unfortunately, no accepted successful second-line chemotherapy currently exists in case of failure of the first line of therapy. Extrapleural pneumonectomy and radiation therapy have resulted in incremental benefit when implemented in the appropriate setting.

The five critical decision making steps in the management of mesothelioma where 2-deoxy-2-[18F]fluoro-d-glucose-(FDG)-positron emission tomography-computed tomography (PET/CT) has been investigated in various research studies include the following: (1) differentiation of MPM from other benign pleural diseases, (2) preoperative staging for selection of operable MPM and post-treatment disease surveillance, (3) evaluation of therapy response, (4) disease prognostication based upon intensity of FDG uptake, and (5) planning of radiotherapy. These are the challenging areas that are encountered by attending physicians. Currently, contrast-enhanced CT is the primary modality for the staging of mesothelioma and for the assessment of therapy response. PET/CT scanning has been shown to have a complementary role in the detection of extrathoracic metastasis; however, recent studies have demonstrated the power of this imaging modality in all phases of the diagnosis and treatment of this cancer in addressing issues related to critical decision making processes.

Search Study and Selection Criteria

References for this review were identified by searches of Medline and PubMed using the terms “PET”, “PET/CT”, “Mesothelioma”, and “FDG” as search terms. The reference lists from all retrieved articles were also assessed for relevant publications. The results of studies, which investigated integrated PET/CT in mesothelioma in at least 15 patients (except for a study investigating radiation therapy planning with PET/CT) were emphasized in this review. Only papers published in English were reviewed without restriction adopted for the publication date. We have restricted the discussion on PET-only literature at selected places in this manuscript where appropriate.

PET/CT in Differentiation of MPM from Benign Pleural Disease

The most common CT findings for MPM are unilateral nodular circumferential pleural and fissural thickening and unilateral pleural effusion usually in association with ipsilateral volume loss of the hemithorax. Other findings include local invasion of the chest wall, mediastinum, diaphragm, or contralateral hemithorax by soft tissue tumor; however, biopsy, particularly in the early stages, is needed to ascertain the nature of the CT findings and to differentiate them from benign causes of pleural thickening and other pleural malignancies, most often due to metastatic disease. Recently, FDG-PET has been suggested as an aid to select the most desirable site within the tumor for biopsy purposes.

In a recently reported study [7], 31 consecutive patients (17 with MPM, nine with benign asbestos-related pleural disease and nine with diffuse pleural fibrosis), underwent FDG-PET/CT to study its utility to differentiate various pathologies. FDG-PET/CT correctly detected the presence of malignancy in 15 of the 17 patients with MPM for a sensitivity, specificity, and overall accuracy of 88.2%, 92.9%, and 90.3%, respectively. FDG-PET/CT correctly identified 13 of the 14 cases of benign pleural disease. The mean standardized uptake value (SUV) was 6.5 ± 3.4 for MPM and 0.8 ± 0.6 for benign pleural diseases (p < 0.001). When the two groups of pleural disease were compared, a cutoff SUV of 2.2 provided the best accuracy with 94.1%, 100%, 100%, and 93.3% sensitivity, specificity, positive predictive value, and negative predictive value, respectively (Table 1).
Table 1

Studies investigating the utility of combined PET-CT imaging in MPM and their salient findings

Study (first author, year)

Clinical setting

No. of patients

Salient findings of the study

Conclusion

Tan C, 2010

Post-treatment surveillance

44

PET/CT had a sensitivity of 94%, a specificity of 100%, and the positive and negative predictive values of 100% and 88%.

FDG-PET/CT is useful in diagnosing disease recurrence after multimodality therapy for MPM.

Yildirim H, 2009

Differentiation of MPM from asbestos-related benign pleural disease

31

A cutoff value of 2.2 for SUV gave the best accuracy with 94.1%, 100%, 100%, and 93.3% for sensitivity, specificity, positive predictive value, and negative predictive value. Mean SUV values were 6.5 ± 3.4 for MPM cases and 0.8 ± 0.6 for benign pleural diseases (p < 0.001).

FDG-PET/CT imaging is a highly accurate and reliable non-invasive test to decide for further investigation of differentiating MPM from benign pleural disease.

Pehlivan B, 2009

PET/CT-based radiation therapy planning

13

In 12 of 13 patients, compared to CT-based delineation, PET/CT-based delineation resulted in a statistically significant decrease in the mean GTV, CTV, PTV1, and PTV2. In these 12 patients, mean GTV decreased by 47.1 ± 28.4%, mean CTV decreased by 38.7 ± 24.7%, mean PTV1 decreased by 31.1 ± 23.1%, and mean PTV2 decreased by 40.0 ± 24.0%. In four of 13 patients, hilar lymph nodes were identified by PET/CT that was not identified by CT alone, changing the nodal status of tumor staging in those patients.

Co-registration of PET and CT information reduces the likelihood of geographic misses, and additionally, significant reductions observed in target volumes may potentially allow escalation of RT dose beyond conventional limits.

Orki A, 2009

Differentiation of benign and malignant pleural diseases

83

PET/CT scanning had 100% sensitivity, 94.8% specificity, and 97.5% accuracy.

PET/CT may prevent redundant surgical procedures in young patients who are SUVmax negative.

Wilcox BE, 2009

Selection of operable malignant pleural mesothelioma

35

PET/CT excluded 14 of 35 patients from surgical intervention. Upstaging from PET/CT occurred in 70% of the patients when surgical pathology was available, two cases to an inoperable stage. Median survival was 20 months for patients undergoing an extrapleural pneumonectomy and 12 months for patients excluded from surgical intervention by integrated PET/CT.

PET/CT is excellent for detecting nodal and distant metastases; however, the ability of this imaging modality to correctly stage locoregional disease is not superior to the combination of CT and MRI.

Veit-Haibach P, 2010

Therapy response evaluation

41

Neither SUVmax response (p = 0.61) nor SUVmean response (p = 0.68) were related to survival. A decrease of tumor lesion glycolysis (TLG) and PETvol, however, was found to be predictive (TLG, p = 0.01; PETvol, p = 0.002).

Response evaluation based on modified RECIST by CT as well as response evaluation by TLG and PETvol in FDG-PET, but not SUVmax measurements are predictive for survival in MPM.

Lee ST, 2009

Prognostic value of FDG-PET/CT

46

Mean SUV(max) of primary pleural lesions in patients with metastatic disease was significantly higher than in patients without metastatic disease (p value < 0.05). Better survival was found in patients without metastases (p value < 0.05).

Detection of extrathoracic metastases on FDG-PET/CT and non-epithelioid histopathology are poor prognostic indicators in patients with MPM.

Sorensen JB, 2008

Preoperative staging

42

Non-curative surgery is avoided in 29% out of 42 MPM patients by preoperative PET/CT and in further 14% by mediastinoscopy.

Even though both procedures are valuable, there are false negative findings with both.

Plathow C, 2008

Limited MPM

54

In stage II, accuracy was 0.77 (CT), 0.86 (PET), 0.8 (MRI), 1.0 (PET/CT), and in stage III 0.75, 0.83, 0.9, 1.0. PET/CT was significantly more accurate (P < 0.05) in stages II and III compared with all other techniques. Interobserver variability was 0.7, 0.9, 0.8, 1.0 in stage II and 0.9, 0.9, 0.9, 1.0 in stage III.

PET/CT stages patients with limited MPM with high accuracy and low interobserver variability.

Krüger S, 2007

Staging changing in MPM

17

Mean SUV was 5.9 ± 1.9 in epithelial MPM and 15.1 ± 10.2 in sarcomatoid MPM. CT and PET/CT revealed discordances in eight out of 17 (47%) patients in TNM classification with four out of eight (50%) being clinically relevant. PET/CT led to downstaging in five (29%) and upstaging in three (18%) patients. Mean survival time tended to be higher in the subgroup of patients with lower mean SUV.

PET/CT seems to be a valuable tool in the staging of MPM and leads to discordant findings in almost every second patient compared to CT alone.

Ambrosini V, 2005

Additional value of PET/CT fusion imaging versus conventional CT scan alone in the staging

15

FDG-PET/CT did not provide additional information about the primary tumor compared to CT, but identified a higher number of metastatic mediastinal lymph nodes in six patients (40% of cases) and unknown metastatic disease to distant sites in three patients (20% of cases). On the basis of PET/CT findings, treatment planning was changed in five patients (33.3% of cases).

FDG-PET/CT fusion imaging can play a relevant role in the staging and treatment planning of MPM patients.

In another study by Orki et al. [8], PET/CT had 100% sensitivity, 94.8% specificity, and 97.5% accuracy for diagnosing MPM. In this study, false positive results were due to pleural tuberculosis. The investigators concluded that PET/CT could be useful as an effective tool for the differentiation of benign and malignant pleural diseases, and therefore may prevent unnecessary surgical procedures (Table 1).

Differentiation of MPM from metastatic adenocarcinoma, however, is difficult with fine needle aspiration or cytology of pleural fluid. A challenge with core biopsy is the choice of biopsy site, as there is usually considerable heterogeneity within the tumor. The number of passes into the tumor for tissue extraction should be minimized since seeding of the biopsy track has been documented and has been shown to be lower with imaging guided biopsy compared to surgical biopsy. FDG-PET/CT guided biopsy can therefore play a critical role in this arena.

FDG-PET and PET/CT in Disease Staging of MPM

Accurate staging of disease is critical for the proper management of patients and has implications for survival. The current tumor, node, metastasis (TNM) staging system for MPM is determined by the recommendations of the International Mesothelioma Interest Group [9]. The critical distinction to be made is differentiation between T3 (resectable) and T4 (technically unresectable) tumors. The latter is characterized by the presence of advanced local invasion of the diaphragm, mediastinum, vital vascular structures, contralateral pleura and chest wall, extensive nodal involvement, and extrathoracic metastatic disease. Those patients that are resectable may be considered for extrapleural pneumonectomy or radical pleurectomy-decortication followed by adjuvant chemotherapy and radiation therapy, which has been associated with improved survival (18–19 months versus 12 months) [10]. The importance of tumor histology as well as tumor resectability and nodal status was introduced by Sugarbaker et al. [10] in an alternative staging system. These investigators demonstrated extended survival in patients who underwent extrapleural pneumonectomy, chemotherapy, and radiation therapy and who had epithelial histology, negative lymph nodes, and negative resection margins.

A distinctive feature of MPM, unlike many other tumors, is its non-spherical, non-compact, and asymmetrical growth pattern that poses a significant challenge to the attending physician not only for the detection of the tumor but also for accurate disease staging and for measurement of response to therapy through conventional anatomical imaging techniques. Contrast-enhanced CT, which is widely regarded as the primary staging tool for MPM, is fairly accurate at differentiating a tumor from a normal tissue [11]; however, this modality has a tendency to underestimate the degree of local invasion, and its accuracy for detecting intrathoracic lymph nodal metastasis is limited [12, 13]. Furthermore, in the post-therapy settings, normal tissue planes may be disrupted, inflamed, or edematous, leading to further difficulty in accurately defining the extent of viable tumor. These limitations often lead to the need for invasive biopsies, which expose the patient to the risk of biopsy track seeding and occasionally to misleading results due to sampling error. The inaccuracy of assessment of the tumor burden and change in the tumor burden has implications for the reliability of the results of phase 2 trials for most experimental therapies for MPM, where response and progression-free survival are the most common and important endpoints [14] to define their efficacy. This has been the prime reason for exploring other modalities to achieve optimal accuracy. Magnetic resonance imaging (MRI) has shown promise for determining the presence of chest wall invasion, pericardial invasion, and transdiaphragmatic extension of tumor, allowing improved prediction of resectability [15]. This is due to the exquisite soft tissue contrast resolution of the MRI and the availability of multiplanar images; however, MRI of the chest can be limited by susceptibility artifacts caused by air/tissue interfaces. MRI also suffers from a reduced sensitivity in the detection of subtle lymph node metastasis, as lymph node abnormalities are assessed based on morphological changes such as nodal size. Therefore, metastatic lymph nodes that are normal in size may escape detection as sites of malignancy. Given the limitations of the structural imaging modalities and the fact that they alone are not sufficient to provide accurate staging information as with other malignancies, FDG-PET as a functional tool has been explored in various settings in MPM.

In early studies, FDG-PET was found to be sensitive for detecting metastatic disease [1619] and is currently implemented as an adjunct to CT to enhance the correct diagnosis of nodal involvement and presence of extrathoracic metastasis. FDG-PET was shown to be superior to CT for differentiating malignant from benign pleural lesions in a study of 28 patients with suspected MPM who underwent PET followed by thoracoscopic or open surgical biopsy (malignant disease was confirmed in 24 and benign disease in four) [17]. When PET is used alone in the assessment of tumor staging, difficulties may arise in determining the exact location of a PET-positive lymph node and in the detection of tumor invasion along the mediastinal and diaphragmatic surfaces. Non-specific uptake by inflammatory, metabolically active, tissues, such as those produced by prior surgery or talc pleurodesis may also result in confounding FDG uptake. As such, most studies have excluded post-surgical/post-pleurodesis PET imaging as an unreliable technique and have elected to wait for a substantial period of time for the inflammation from the therapeutic intervention to subside. For the talc pleurodesis, the assumed period of time for the non-specific FDG uptake has been stated to be around 6 months. PET/CT, which combines low-dose CT and PET scan in the same examination, provides superior anatomical localization of the radiotracer uptake and thereby enhances the accuracy of the PET scanning technique with only a small incremental increase in radiation delivered to the patient. The fused imaging increases the specificity of PET and the sensitivity of CT and, therefore, is superior to either scan alone. Also, the CT portion provides the necessary tissue attenuation correction needed to calibrate radiotracer uptake, previously accomplished by an external radioactive source used during the PET and also results in scan completion in a shorter time period. While the combined modality provides many potential benefits, there are certain technical considerations with regard to the optimal use of this technique. Beam hardening artifacts caused by contrast agents (this is not used routinely in most labs) and metallic hardware are due to the differences in the X-ray energy (100 keV) and gamma ray energy (511 keV) of photons emitted from FDG. Therefore, contrast-enhanced CT should not be used for attenuation correction purposes, and administered oral contrast agents should be diluted. In addition, FDG uptake may be falsely elevated near metallic hardware, necessitating careful interpretation in these areas [20]. In the chest, there is the additional problem of respiratory motion artifact since the CT examination is usually obtained during breath holding, whereas the PET scan is acquired during free breathing. This not only affects the spatial resolution of PET images but also the accuracy of the attenuation correction information provided by the CT scan. While this mismatch due to respiratory motion can be a potential source of error, it does not appear to have a significant effect on the diagnostic quality of the images acquired [21]. PET/CT has proven to be more accurate in the staging of other thoracic malignancies than either modality alone [22]. Combined imaging has demonstrated accurate delineation of chest wall invasion, differentiation of tumor from atelectasis, guidance of mediastinoscopy for precise sampling of the lymph nodes for accurate staging, and localization of suspected extrathoracic metastasis [23].

PET/CT in Preoperative Disease Staging of MPM

Similar to FDG-PET, FDG-PET/CT has also shown to be of great value in determining resectability of MPM [2428] and has been investigated by several groups across the world (Table 1).

In a comparative study by Plathow et al. [25], the accuracy of CT, PET, MRI, and PET/CT was 0.77, 0.86, 0.8, and 1.0, respectively for limited MPM patients, and 0.75, 0.83, 0.9, and 1.0, respectively for patients in stage II and III. CT and MRI were not able to detect distant metastases in two patients, which changed therapy (operable versus inoperable). The authors concluded FDG-PET/CT was significantly more accurate (p < 0.05) in stages II and III compared with all other techniques. In another study by Sorensen et al. [26] comparing the accuracy of preoperative staging with different imaging modalities, it was found that non-curative surgery could be avoided in 29% of 42 MPM patients by preoperative PET/CT and in another 14% by employing mediastinoscopy. The authors concluded that both procedures are valuable for the accurate staging of MPM (Table 1).

In a retrospective report [27], the utility of PET/CT in the initial staging was assessed in 35 patients with proven MPM. PET/CT excluded 14 of 35 patients from surgical intervention. Upstaging from PET/CT was noted in 70% of the patients when surgical pathology was available. Median survival was 20 months for patients undergoing an extrapleural pneumonectomy and 12 months for patients excluded from surgical intervention by PET/CT. The authors concluded that PET/CT is excellent for detecting nodal and distant metastases. Ambrosini et al. had examined the additional value of PET/CT versus conventional CT alone in the staging of MPM. FDG-PET/CT did not provide additional information about the primary tumor compared to CT, but identified a higher number of metastatic mediastinal lymph nodes in six patients (40% of cases) and unknown metastatic disease to distant sites in three patients (20% of cases). On the basis of the PET/CT findings, treatment planning was changed in five patients (33.3% of cases). Similarly, it can also reduce the apparent tumor volume that needs to be irradiated when it downgrades the tumor stage.

Overall, from the evidence obtained from these studies, it appears that PET/CT increases the accuracy of staging in patients with MPM and improves the selection of patients for curative surgical resection. It is particularly useful for identifying occult distant metastases.

PET/CT in Post-Treatment Disease Surveillance of MPM

In a recent retrospective report [29], PET/CT was performed after multimodality therapy when disease recurrence was suspected. Overall, PET/CT had a sensitivity of 94%, a specificity of 100%, and the positive and negative predictive values of 100% and 88%, respectively. The authors concluded that PET/CT is useful in diagnosing disease recurrence after multimodality therapy for MPM (Table 1).

Response to Chemotherapy/Radiation

An emerging role for FDG-PET/CT is that of therapeutic response assessment. This has been recently shown to hold promise not only for MPM but also for a variety of other malignancies including lymphoma, lung cancer, and breast cancer [30, 31]. In MPM management, a measureable response or lack thereof is not usually detectable by CT until after multiple cycles of chemotherapy. This is likely due to the insinuating morphology and asymmetric growth of the tumor which create significant challenges for accurate measurement of tumor burden on CT. In order to address the problem of non-spherical morphology of MPM, Byrne et al. have proposed a modified response evaluation criteria for solid tumors (RECIST) which may be applicable to this tumor [32]. In this scheme, the overall tumor burden is assessed by summing a total of six measurements of the tumor maximal thickness perpendicular to the chest wall, with two measurements at reproducible landmarks carried out on three different slices at least 1 cm apart. In this study, response according to the modified criteria predicted for superior survival (15.1 versus 8.9 months; p = 0.03) and forced vital capacity increase during treatment (p < 0.0001). While the modified RECIST represents an improvement in the accuracy of response assessment, it still fell short in depicting the true representation of total tumor burden and subtle growth extension in non-axial planes such as along the pleural fissures. A study of 22 patients with MPM demonstrated significant interobserver variability in tumor measurement (ranging up to 30% difference) using modified RECIST. In addition to limitations in the assessment of overall tumor growth, CT also suffers in sensitivity to detect early tumor extension across the diaphragm, involvement of regional lymph nodes that are not yet large enough to be considered abnormal, and subtle invasion into the chest wall.

Early evidence suggests that PET/CT may be able to detect a response to therapy before there is a measureable morphological change on CT. In a study of 41 patients with MPM [33], a decrease in total glycolytic volume (TGV) after chemotherapy was associated (p < 0.015) with improved survival. CT response after three cycles of chemotherapy was significantly related to overall survival (p = 0.001). PET measurements, using SUVmax or SUVmean did not significantly correlate with survival [33]. These investigators demonstrated that a drop in tumor lesion glycolysis or TGV, as measured on FDG-PET/CT, obtained after three cycles of chemotherapy was predictive of a therapeutic response as determined by RECIST [33]. Measurements of tumor PET uptake may be more accurate when obtaining volumetric measurements as opposed to SUVmax. In another study, 20 MPM were assessed before and after two cycles of chemotherapy with CT and PET. A 25% decrease in SUVmax correlated with improved time to progression (14 months versus 7 months). Tumor changes on CT did not correlate with tumor response [34]. Therefore, early evidence suggests that FDG-PET may have a useful role in evaluating MPM response to therapy. In one study [35, 36], evidence of response was reported as early as after one cycle of chemotherapy by quantitative semi-automated volume-based FDG-PET analysis performed to obtain the TGV of tumor (Table 1).

Prognosis Defined by FDG Uptake of Tumor

The tumor avidity for FDG, as measured by the SUV, is also now regarded as a surrogate marker of tumor biology, and hence carries prognostic significance. Data from various studies suggest that tumor SUVmax from FDG-PET may serve as a prognostic marker in MPM along with variations in the histologic subtype, mediastinal nodal status, presence of extrathoracic metastasis, and elevated tumor growth markers such as vascular endothelial growth factor expression. This is important as prognostic stratification has implications for clinical management of patients and appropriate enrollment of subjects into clinical trials.

Flores et al. examined this in a population of 137 patients with MPM [17] and observed that FDG-PET tumor uptake greater than SUVmax of ten predicted a median survival of 9 months versus 21 months for those with SUVmax less than ten (p = 0.02). In another study by Benard et al. [53], there was a suggestion that initial SUVmax may correlate with survival. In their study, there was a negative linear correlation between initial tumor SUVmax and survival with a significantly decreased survival noted in those patients with SUVmax greater than four.

Disease prognostications with PET/CT in terms of overall survival have been examined in three studies. In a retrospective review [37] of FDG-PET/CT performed in patients with biopsy-proven MPM, mean SUVmax of primary MPM in patients with metastatic disease was significantly higher than in patients with MPM without metastatic disease (p < 0.05). Longer survival was noted in patients without metastases (p < 0.05). The study results indicated that detection of extrathoracic metastases on FDG-PET/CT and non-epithelioid histopathology are poor prognostic indicators in patients with MPM. In a study by Kruger et al. [38], SUVmean was 5.9 ± 1.9 in epithelial MPM and 15.1 ± 10.2 in sarcomatoid MPM. The mean survival time tended to be higher in the subgroup of patients with lower SUVmean. In a study by Wilcox et al. [27], the median survival was 20 months for patients undergoing an extrapleural pneumonectomy and 12 months for patients excluded from surgical intervention by PET/CT (Table 1).

PET/CT for RT Planning in MPM

In order to compare CT with PET/CT as the basis for delineating target volumes in these patients, Pehlivan et al. [39] retrospectively evaluated data from these two approaches in 13 patients with histologically proven MPM. For each patient, target volumes [gross tumor volume (GTV), clinical target volume (CTV), and planning target volume (PTV)] were defined using the required data sets. In 12 of 13 patients, comparing CT-based delineation to that of PET/CT-based delineation resulted in a statistically significant decrease in the mean GTV, CTV, PTV1, and PTV2. In these 12 patients, mean GTV decreased by 47.1 ± 28.4%, mean CTV decreased by 38.7 ± 24.7%, mean PTV1 decreased by 31.1 ± 23.1%, and mean PTV2 decreased by 40.0 ± 24.0%. In four of 13 patients, hilar lymph nodes that were not identified by CT alone were identified by PET/CT, changing the nodal status of tumor staging substantially in these patients. The investigators concluded that co-registration of PET and CT information reduces the likelihood of geographic misses. In addition, significant reductions observed in target volumes may potentially allow escalation of RT dose beyond conventional limits (Table 1).

Response to Novel Therapeutic Regimens: Promise for Drug Development

Given the dismal success rates of the available therapeutic interventions, novel first-line and second-line treatments are urgently needed. New therapy candidates on the horizon for treating MPM include anti-angiogenic agents, gene therapy, and antifolates. One major obstacle that hinders the progress of development of the novel therapy and optimization of current therapies is the lack of accurate diagnostic modalities for the staging and measurement of therapy response. Currently, determination of response to novel biological therapies for MPM including vaccine and gene therapy trials [4042] is of great clinical importance. The benchmarks for assessing post-therapy response on PET have not been well delineated in these domains; however, it is conceivable that molecular imaging with PET will likely play a pivotal role in this regard and thus aid in the development of appropriate protocols for this aggressive malignancy.

Dual Time Point PET Imaging in Disease Characterization

In a recently reported study [54], consecutive patients with suspected and/or recurrent MPM underwent two sequential FDG-PET scans (dual time point imaging). Patients were divided into three groups (A = newly diagnosed MPM, B = recurrent MPM, and C = benign pleural disease). The parameters of FDG uptake (SUVmax and its change over time) were compared among the groups. Of this population examined, 44 were diagnosed with MPM (28 newly diagnosed and 16 recurrent). The PET studies demonstrated 229 malignant pleural lesions in these patients. The remaining 11 patients were proven to have benign pleural disease. The mean ± SD of the SUV(max1), SUV(max2), and the Delta%SUV(max) of all the lesions of each patient in groups A, B, and C were 5.0 ± 2.2%, 5.8 ± 2.8%, and 12.8 ± 8.4%; 4.6 ± 1.7%, 5.3 ± 2.0%, 13.8 ± 9.2%; and 1.6 ± 0.4%, 1.4 ± 0.3%, and −9.6 ± 19.1%, respectively. The mean ± SD of the SUV(max1), SUV(max2), and Delta%SUV(max) in patients with both newly diagnosed and recurrent MPM were significantly higher than those of the benign pleural disease group (p < 0.0001). The authors concluded that there is an increasing uptake of FDG over time in pleural malignancies, whereas the uptake in benign pleural disease generally stays stable or decreases with time. Therefore, dual time point imaging appeared to be an effective approach in differentiating benign from malignant pleural disease.

Novel PET Radiotracers on the Horizon: Imaging Tumor Proliferation and Hypoxia

Novel PET radiotracers are increasingly employed for the diagnosis and management of a variety of malignancies at different stages of therapeutic interventions. One such compound is 18Fluorine-labeled thymidine (FLT) which is a fluorinated analogue of thymidine. Multiple studies have demonstrated a close correlation between proliferative rate and FLT-PET uptake [43]. In fact, FLT uptake is quite specific for tumor proliferative rate (as measured by Ki67 staining), and thus may have a role in the differentiation between tumor and benign inflammatory processes which can be difficult to differentiate, particularly in the post-operative and post-therapeutic settings. FLT-PET may also have a role in the assessment of tumor response to therapy [43]. FLT-PET is of interest in the management of MPM, particularly if this tracer continues to prove to be specific for malignancy. This approach can find important use in the post-therapy setting (including pleurodesis, surgery, and radiation therapy), where non-specific FDG accumulation has been reported. In these settings, benign inflammation related to therapy may lead to inappropriate interpretation of FDG-PET. In the case of pleurodesis, benign inflammatory increased uptake can be seen months after therapy. FLT-PET also requires investigation in the setting of MPM for the determination of early response to therapy. In this patient population with a poor prognosis, an early switch to an effective therapy may have a significant impact on patient outcome. Tumor hypoxia has been widely recognized to lead to an aggressive phenotype and relative resistance to both radiation therapy and/or chemotherapy [44, 45]. Available PET hypoxia-specific agents can potentially provide evidence for this biologically important phenomenon. Since hypoxic tumor may be relatively more resistant to therapy, it would be useful to visualize specific sites of hypoxia within the tumor so that the therapy may be tailored accordingly. This will be particularly relevant to radiation therapy planning in this cancer.

The data, with regard to these and other novel PET agents in MPM, are limited and primarily deal with preclinical animal studies [46]. In a recent report using mouse models, it was observed that FLT and methyl-[(11)C]thiothymidine [(11)C]S-dThd may be suitable for the epithelioid subtype, and FDG seemed suitable for the sarcomatoid subtype of MPM. The cellular uptake of [(14)C]FDG and [(3)H]FLT and thymidine kinase 1 activity in sarcomatoid cells were higher than those of epithelioid cells.

Novel Approach of WBMB Estimation in Mesothelioma

Our group at the University of Pennsylvania has performed a global disease assessment by combined structure (CT-MRI) and function (PET) in a variety of disorders for the past two decades [4752]. Recently, we have employed a new method of image segmentation based on an iterative thresholding algorithm (region of interest visualization, evaluation, and image registration (ROVER) ABX, Radeberg, Germany [52]), which permits definition of the boundaries of lesions based on PET images alone to provide lesional metabolically active tumor volumes, lesional partial volume corrected SUV (PVC-SUV) measurements, lesional PVC metabolic burden (PVC-MB) (calculated as the product of lesional MVP and lesional PVC-SUV), and whole body metabolic burden (WBMB) (calculated as the sum of lesional PVC-MB of all lesions). This approach substantially improves the ability to calculate the disease burden by taking into consideration both volumetric and metabolic characteristics of the disease rather than relying on these quantitative parameters independently. Such an approach provides an efficient means for quantifying widespread disease processes in a reproducible and accurate manner and has great potential for the objective assessment of response following therapeutic intervention. Our preliminary data from this approach, which are shown in Fig. 1, show feasibility of this method for assessing disease activity in this particular malignancy. By using this approach, we will be able to better characterize the nature of the disease process, to assess patient prognosis, and to assess for treatment response.
https://static-content.springer.com/image/art%3A10.1007%2Fs11307-010-0426-6/MediaObjects/11307_2010_426_Fig1_HTML.gif
Fig. 1

PET images in coronal (a), sagittal (b), and transverse (c) planes as part of FDG-PET/CT study in a 78-year-old male with malignant pleural mesothelioma demonstrate FDG uptake within left pleura. Same PET images with color overlay following use of ROVER software reveal accurate segmentation of pleural disease. Another coronal (d) and sagittal (e) section of this image.

Conclusion

The use of combined PET/CT imaging in MPM appears to provide optimal disease characterization with a high negative predictive value for the diagnosis, superior disease staging both at initial diagnosis and later in the disease course, better monitoring of treatment response, and accurate assessment of target volume for radiotherapy planning. It is imperative that this promising molecular imaging tool be employed more extensively by investigators to develop new and more effective therapeutic approaches in the future to improve the overall outcome of patients with this grave malignancy.

Conflict of Interest Statement

The authors declare no conflict of interest.

Copyright information

© Academy of Molecular Imaging and Society for Molecular Imaging 2010