computed tomography and multidetector computed tomography roles in thymic lesion treatment planning

Lung Cancer (2008) 61, 362—368

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/lungcan

[18F]FDG positron emission tomography/computed tomography and multidetector computed tomography roles in thymic lesion treatment planning Laura L. Travaini a, Giuseppe Petralia b, Giuseppe Trifir` o a, Laura Ravasi c, Domenico Galetta d, Giuseppe Carbone a, Fabio Falcini e, Lorenzo Spaggiari d,f, Massimo Bellomi b,f, Giovanni Paganelli a,∗ a

Division of Nuclear Medicine, European Institute of Oncology, 20141 Milan, Italy Division of Radiology, European Institute of Oncology, Milan, Italy c Nuclear Medicine Resident, University of Milan, Italy d Division of Thoracic Surgery, European Institute of Oncology, Milan, Italy e Division of Medical Oncology, IRST Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori, Meldola (FC), Italy f School of Medicine, University of Milan, Italy b

Received 17 September 2007; received in revised form 2 November 2007; accepted 24 January 2008

KEYWORDS Epithelial thymic lesions; Fluorodeoxyglucose; PET/CT; Multidetector CT; Benign thymic lesions; SUV

∗

Summary Rationale: Thymic masses may represent an unsolved diagnostic problem which often require surgical procedures for an accurate staging. A non-invasive way to determine the nature of thymic lesions would help identify the patients which are true candidates for surgery. Our retrospective study aims to assess multidetector computed tomography and 2-[18 F]fluoro-2deoxyglucose positron emission tomography/computed tomography ([18 F]FDG-PET/CT) capacity to distinguish benign from malignant thymic lesions. Methods: Helical multidetector CT (MDCT) and [18 F]FDG-PET/CT of twenty consecutive patients presenting with a thymic mass at our Institute were retrospectively analyzed. MDCT scans were focused on morphologic features and invasiveness characteristics. Qualitative and semiquantitative analyses by maximum standardized uptake value corrected for body weight (SUVbw max) were performed on [18 F]FDG-PET/CT. In all cases, readers were blinded to pathology findings. Both imaging techniques were correlated to final pathology. Student’s t-test was performed on SUVbw max stratified for thymic epithelial tumors. Results: In the group of benign lesions MDCT correctly identified well-defined margins of masses in 8 out of 8 patients whereas [18 F]FDG-PET/CT was negative in 7 out of 8 patients.

Corresponding author. Tel.: +39 02 57489043; fax: +39 02 94379223. E-mail address: [email protected] (G. Paganelli).

0169-5002/$ — see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.lungcan.2008.01.019

[18 F]FDG-PET/CT and MDCT in thymic lesions

363

Among malignant lesions MDCT revealed mediastinum fat or infiltration of adjacent organs in 10/12 patients. On the other hand [18 F]FDG-PET/CT showed increased radiotracer uptake in 12/12 patients. Conclusions: MDCT and [18 F]FDG-PET/CT alone are not able to differentiate the nature of thymic lesions. However, they are two non-invasive complementary techniques which can be used to differentiate benign from high-risk malignant thymic lesions. These findings should be taken into account before surgery is performed as a diagnostic procedure. © 2008 Elsevier Ireland Ltd. All rights reserved.

1. Introduction The thymus is a small organ that lies in the upper chest under the breastbone. It belongs to the lymphatic system and undergoes a physiologic involution from adolescence to adulthood. Physiologically throughout childhood or under special conditions [1—3] the thymus can be hyperplastic, regardless of age. Follicular or lymphoid hyperplasia of the thymus is frequently found in association with autoimmune disorders, especially in patients with myasthenia gravis. The most common masses in the anterior compartment are of thymic, lymphatic, or germ cell origin; thymic is the most common. Benign thymic lesions are commonly represented by multilocular cysts that may occur de novo or may be associated with a variety of systemic and localized mediastinal disorders [4]. Malignant lesions are mainly represented by epithelial and non-epithelial tumors [5]. In adults thymic tumors have been described as the second most common type of mediastinal tumor. When considering all age groups, nearly 55% of patients with benign mediastinal masses are asymptomatic at presentation, compared to only approximately 15% of those in whom masses are found to be malignant. Thymoma is a rare neoplasm of thymic epithelial cells that accounts for 0.2—1.5% of all malignancies and 0.06% of all thymic cancers. It can be solid, cystic, with smooth or rough margins, limited or diffusely invasive, soft or with calcifications. Six out of ten are capsulated and the remainder is invasive. Both invasive and non-invasive thymomas show cytological characteristics of benignancy. The most common systemic manifestation associated with thymoma is myasthenia gravis. This occurs in 10—50% of patients with thymoma and is thought by many to be an autoimmune phenomenon [6]. In contrast, thymic carcinomas are locally invasive, with a high risk of relapse and death [7] and, unlike thymoma, frequently metastasize to regional lymph nodes and distant sites [8]. The nosology of thymic epithelial tumors is somewhat confusing. Thymic epithelial tumors consist of several histological subtypes: thymoma A, AB, B1, B2, B3 and thymic carcinoma in order of increasing malignancy [9]. Thymomas A and AB are generally indolent tumors without invasiveness features. B1—B3 thymomas are however characterized by an increased malignancy. The age peak of incidence is between 40 and 60 years old. One out of two patients is diagnosed with thymomas by chance under routine chest X-ray. Most thymomas are diagnosed and staged at surgical intervention that often consists of an en-bloc resection. In fact surgical resection is

offered to patients bearing a mediastinum mass suspected (though not yet confirmed) to be a thymoma [10]. A noninvasive methodology that accurately determines the nature of the thymic mass would benefit the patients bearing benign thymic lesions by sparing them from aggressive diagnostic procedures. Contrast-enhanced computed tomography (CT) [11] helps characterizing thymomas and thymic carcinomas by distinguishing those from benign mediastinum tumors (e.g. cysts, hyperplasia) or other neoplasia (e.g. lymphoma, non-small cell lung cancer, extra-gonadic seminoma). However, several primary mediastinum masses share the same morphologic imaging features and resection is often advisable for definitive characterization [12]. On the other hand Positron Emission Tomography (PET) detects increased glucose uptake in malignant tissue using 2-[18 F]fluoro-2-deoxyglucose ([18 F]FDG) as radiotracer. The role of [18 F]FDG-PET/CT has already been recognized and validated for tumor staging of various organs, e.g. lung [13], stomach and intestine [14,15] and head and neck cancer [16]. In thymic lesions, the role of [18 F]FDG-PET/CT has been partially investigated [17—21]. As part of our continuous effort to look for the least invasive accurate diagnostic methodology, we hereby assessed the roles of helical multidetector CT (MDCT) and [18 F]FDGPET/CT in determining the nature of thymic masses.

2. Materials and methods 2.1. Patients We retrospectively analyzed MDCT and [18 F]FDG-PET/CT images of 11 males and 9 females (average age 55 years old; range 24—78) referred to our Institute for a indeterminate thymic mass. Based on pathology, thymic lesions were differentiated in benign or malignant and those further classified according to the World Health Organization (WHO) classification [9] and staged according to the Masaoka system [22]. All malignant lesions were epithelial thymic tumors and were grouped into 3 subgroups: low-risk (types A, AB and B1), high-risk thymoma (types B2 and B3) and thymic carcinomas, as previously described by Sung et al. [23].

2.2. Chest MDCT All exams were performed on a 16-slice scanner (Lightspeed 16, GE Medical Systems, Milwaukee, WI, USA). The following parameters were used: 2.5 mm-thickness contiguous

364 slices with standard reconstruction filter, 100—300 mA (automatic), 120 kV, rotation time 0.8 s, with 50 s scanning delay after injection of 350 mgI/ml contrast medium (2 ml/kg), followed by 20 ml saline injection, at 2 ml/s flow rate, via a 18—20 gauge cannula in the antecubital vein. MDCT images were analyzed on Advantage AW4.2 workstation with multiplanar reformat capability. For research purposes, a single reader with 4 years of experience in oncology imaging performed a second reading of the scans; this reader was no longer blind to the pathology results. Thymic mass margins were accurately evaluated. Welldefined margins and homogeneous aspect of adjacent mediastinal fat or adjacent organs were considered suggestive for non-infiltrative thymic masses, thus read as benign. On the other hand, if thymic masses were characterized by one or more of the following criteria (i.e. ill-defined margins, solid strands into mediastinal fat, loss of fat planes, loss of solid strands into adjacent organs) mediastinal fat or adjacent organ infiltration was documented therefore lesion was MDCT-read as malignant. Additional lesion features such as maximum transverse diameter, its morphology (regular or lobulated), the presence of calcifications and the homogeneity of the structure (homogeneous and non-homogeneous) were reported for all thymic lesions and compared to pathology results. The results were classified as true positive (TP), true negative (TN), false positive (FP) and false negative (FN). The evaluation of the results was based on calculation of sensitivity (TP/TP + FN), specificity (TN/TN + FP), positive predictive value (TP/TP + FP) and negative predictive value (TN/TN + FN).

2.3. [18 F]FDG-PET/CT The isotope and the radiotracer were produced and synthesized, as previously reported [24]. After fasting for about 6 h, patients were intravenously administered with 5 MBq/kg [18 F]FDG. Blood glucose levels, measured right before radiotracer injection, were below 150 mg/dl. No patients suffered from diabetes mellitus. Fifty minutes after radiotracer injection, images were acquired with a dedicated PET/CT system in 2D mode. PET/CT device (Discovery LS, GE Medical Systems, Waukesha, WI) consisting of an Advance Nxi PET scanner and an eight-slice Light Speed Plus CT scanner. The axes of both systems were mechanically aligned so that shifting the examination table by 60 cm moved the patient from the CT into the PET gantry. The resulting PET and CT images were co-registered on hardware. Patients were positioned head-first supine, and moved to just above the first scanning position on the CT. A scout-scan was acquired to define the axial imaging range, which for a whole-body PET/CT examination typically extended from the lower jaw to the upper thighs. The standard protocol for CT image consisted of 120 kev, 80 mA, a tube-rotating time of 0.8 s per rotation, a pitch of 1.5, and a TC section thickness of 5 mm, which was matched to the section thickness of the PET images (4.25 mm). Immediately after CT, PET covered the identical axial field of view. The acquisition time for PET was 4 min per table position

L.L. Travaini et al. and 24—28 min in all. Patients were instructed to breathe normally. PET-image data sets were reconstructed iteratively with segmented correction for attenuation with use of the CT data. Co-registered images were displayed by means of Xeleris software (GE Medical Systems). PET images were analyzed qualitatively and semiquantitatively by maximum standardized uptake value corrected for body weight (SUVbw max) by three independent readers with 4 years of experience in oncological field, kept blind to patient’s history and thymic lesion pathology results. In the mediastinum qualitative analysis outcome was reported as negative if no uptake was visible, positive if uptake was greater than that of the physiological mediastinum or doubtful if uptake was similar to the one of physiological mediastinum. Semi-quantitative analysis, through SUVbw max, was carried out by one of the three experienced physicians, by drawing a region of interest on focal [18 F]FDG uptake in the mediastinum, when present, or on the whole mediastinum region, when qualitative analysis was negative. Standardized uptake values were grouped according to pathology results. Student’s t-test was performed on benign versus low-risk lesions, benign versus high-risk lesions, low versus high-risk lesions and benign versus all malignant ones. Inter-observer agreement was reported. When discordant interpretations were given, images were re-analyzed and stratified from 0 to 2 as 0: no uptake; 1: uptake similar to mediastinum; 2: greater uptake than mediastinum. As [18 F]FDG-PET/CT is a whole-body imaging methodology, non-thymic abnormal radiotracer uptake was further investigated when revealed. Similarly to MDCT analysis, sensitivity, specificity and predictive values were calculated for [18 F]FDG-PET/CT.

2.4. Pathology Surgical specimen of all thymic lesions were analyzed. Malignant lesions were staged according to WHO classification and Masaoka system. If MDCT and [18 F]FDG-PET/CT detected additional lesions, these were submitted to cytology and/or histo-pathology analyses to define their nature.

3. Results Characteristics of thymic lesions by MDCT and [18 F]FDGPET/CT and post-surgical pathology results are reported in Table 1. Among benign lesions, MDCT correctly identified welldefined margins of thymic masses in all cases, indicating eight benign lesions (specificity 100%). Eleven out of 12 malignant thymic masses were classified as stage ≥2 according to Masaoka because of their invasiveness. MDCT identified mediastinum fat or adjacent organ infiltration in 9 out of 11 patients: it failed to assess infiltration of mediastinum fat in case #12 and lung parenchyma in case #19. MDCT showed 81.8% sensitivity for predicting infiltration in mediastinum fat or adjacent organs (instead of the

Case #

Age (years)

Type of risk

MDCT analysis

[18 F]FDG-PET/CT analysis

Pathology classification

Max diameter (mm)

Morphology

Calcifications

Structure

Margins

Reading outcome

Qualitative

Standard uptake value

World Health Organization

Masaoka

Follow up

Cystis Cystis Cystis Thymus Hyperplasia Hyperplasia Hyperplasia Hyperplasia B1 AB AB AB B1 AB B2 B2/B3 B3 B2 Thymic carcinoma Thymic carcinoma

— — — — — — — — II II II II I II II II IVB IVA III

None None None None None None None None RT RT RT MDCT RT MDCT RT RT RT RT RT

II

CT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

68 69 58 48 24 53 32 74 54 65 45 38 51 78 34 58 58 71 47

Benign Benign Benign Benign Benign Benign Benign Benign Low risk Low risk Low risk Low risk Low risk Low risk High risk High risk High risk High risk Carcinoma

28 47 27 12 12 25 18 48 15 93 54 69 46 78 53 22 55 60 78

Regular Regular Regular Regular Regular Regular Regular Regular Lobular Lobular Regular Lobular Regular Regular Lobular Lobular Lobular Lobular Regular

No No No No No No No No No No No No Yes No No No Yes No No

h h h h h h h n-h n-h n-h h n-h n-h n-h n-h n-h n-h n-h h

Well-defined Well-defined Well-defined Well-defined Well-defined Well-defined Well-defined Well-defined Infiltrative Infiltrative Infiltrative Well-defined Well-defined Infiltrative Infiltrative Infiltrative Infiltrative Infiltrative Well-defined

b b b b b b b b m m m b b m m m m m b

b b b b b b b m m m m m m m m m m m m

0.9 1.3 2.2 2.2 1.7 2.8 3.2 5.0 2.3 2.7 2.9 4.0 6.5 15.5 5.0 6.9 7.0 7.2 8.0

20

67

Carcinoma

63

Lobular

No

h

Infiltrative

m

m

9.5

[18 F]FDG-PET/CT and MDCT in thymic lesions

Characteristics of thymic lesions by MDCT, [18 F]FDG-PET/CT and pathology analyses

Table 1

MDCT: multidetector computed tomography; [18 F]FDG-PET/CT: positron emission tomography/computed tomography with fluorodeoxyglucose; h: homogeneous; n-h: non-homogeneous; b: benign; m: malignant; RT: 50.4 Gy of radiotherapy; CT: chemotherapy.

365

366

Fig. 1 [18 F]FDG-PET/CT semi-quantitative analysis by standard uptake value (SUV) of thymic lesions according to type of risk.

PPV and NPV relative to malignant or benign nature of the lesion): PPV and NPV was, 100% and 80%, respectively. All benign thymic lesions were smaller than 5 cm, whereas 9 out of 12 malignant ones were greater than 5 cm. Moreover, all benign thymic lesions presented at MDCT regular morphology, no calcifications and 7 out of 8 homogeneous structure, whereas among the 12 epithelial thymic tumors 8 presented with lobulated morphology, 2 with calcifications and 9 with non-homogeneous structure. From [18 F]FDG-PET/CT readings, six scored negative, twelve scored positive and two were discordant: re-analyses of these two gave positive score in case #11 and negative score in case #7. Among the eight benign lesions, PET was interpreted as negative in seven cases with a SUV (mean ± standard deviation) of 2.0 ± 0.8 (specificity 87.5%). The benign lesion (case #8) interpreted as malignant had SUV equal to 5.0. [18 F]FDG-PET/CT exams were positively interpreted in 13 cases and mean ± standard deviation of SUV was 6.3 ± 3.5: PPV and NPV was 92.3% and 100%, respectively. In the group of epithelial thymic tumors, analysis according to radiotracer uptake revealed 12/12 positive results (sensitivity 100%). According to pathology results, the lesions were grouped in benign, low risk and high risk. SUVs ± standard deviation were 2.42 ± 1.29 (n = 8), 5.64 ± 5.04 (n = 6), 5.86 ± 1.02 (n = 4), respectively. Cases #19 and 20, classified as thymic carcinoma, had SUVs of 8.0 and 9.5. Student’s t-test was significantly different when benign versus all malignant (p = 0.003) and benign versus high-risk group (p = 0.0001) were compared. Other comparisons did not reach statistical significance even though a trend of increasing SUV was present along with increasing malignancy (Fig. 1). In addition to thymic epithelial tumors, [18 F]FDG-PET/CT detected focal uptake at neck level in case #18. Further work up by MDCT and cytology documented a Warthin neoplasia.

4. Discussion Since surgical intervention is internationally recommended as the treatment of choice for a majority of primary mediastinal mass lesions, be they malignant or benign, we intended to increase justification to spare invasive diagnostic treatment to patients bearing mediastinum masses.

L.L. Travaini et al. Particularly we assessed accuracy of non-invasive diagnostic tools as MDCT scanning and [18 F]FDG-PET/CT in differentiation of benign from malignant thymic lesions. Variable results have been reported in the literature for MDCT assessment of thymic lesions. Jung et al. [25] reported significantly smaller diameter for atypical thymoma, compared to thymic carcinoma, but not significant differences between the two histology types in morphologic features. Several morphologic features in CT images have been correlated with the different histological subtypes of thymic mass. Necrosis, hemorrhage, cystic degeneration or calcifications are mostly seen more in thymic carcinoma and invasion of great vessels, lymph node enlargement, extrathymic metastases may differentiate thymic carcinoma from other histology subtypes. Large size of the tumor, multifocal calcifications and necrotic foci are characteristics more commonly seen with invasive thymomas [26,27]. Our results are concordant with previous reported experiences, since all benign lesions were smaller than 5 cm, presented regular morphology and no calcifications. However, among the 12 malignant thymic lesions, 3 were smaller than 5 cm, 4 presented with regular morphology and homogenous structure and 10 presented no calcifications. In our cohort, then, maximum diameter greater than 5 cm and lobulated morphology seemed to better predict malignancy, but no statistical power could be achieved with our small study cohort and larger studies are required to validate these initial results since reliable differentiation between histological subtypes of thymic epithelial tumors is not currently possible with MDCT. MDCT reached 100% specificity by showing infiltration of mediastinum fat or adjacent organs in 10 lesions surgically excised and pathologically classified as malignant. MDCT sensitivity dropped to 81.8% because it failed to identify mediastinum fat and lung parenchima infiltration in 2 patients, although it correctly identified as benign the false positive [18 F]FDG-PET/CT (case #8). Although MDCT has shown a high accuracy in discriminating the lesions according to the presence of infiltration, some malignant lesions may not present features of infiltration and can be falsely interpreted as benign, relying on MDCT alone. This was case #13, an encapsulated B1 thymoma that would have been classified as benign, if only MDCT assessment of infiltration characteristics had been considered. Thus, absence of morphologic MDCT findings suggestive for invasive thymoma or thymic carcinoma cannot exclude invasive procedures (i.e. biopsy and surgery), considered the gold standard for diagnostic purposes. As far as [18 F]FDG-PET/CT scans, there was a 100% interreader agreement and qualitative analyses went along with semi-quantitative ones. In fact, [18 F]FDG-PET/CT true negative results matched SUVs < 2.2. Such value seems to be a threshold beneath which most benign thymic lesions lay. However, one benign lesion was interpreted as malignant: it was case #8, a 74-year-old man with articular rheumatism. Similarly, case #7, a 32-year-old woman, had a thymic hyperplasia secondary to chemotherapy for gynecological cancer (Fig. 2). Both situations may increase thymic uptake as reported by Tatebe et al. [3] and Halaweh et al. [27]. Had the physicians been aware of the clinical history of these two patients, exam interpretation could have been

[18 F]FDG-PET/CT and MDCT in thymic lesions

367

Fig. 2 Axial views of mediastinum through CT scan (a), MDCT (b) and [18 F]FDG-PET/CT fused image (c) of a 32-year-old woman (case #7) bearing thymic hyperplasia. SUV of ROI is 3.2. Axial MDCT image showed well-defined margins of the thymic mass, indicating no adjacent organs infiltration.

Fig. 3 Axial views of mediastinum through [18 F]FDG-PET/CT scan (a), MDCT (b) and FDG PET/CT fused image (c) of a 38-year-old man (case # 12). [18 F]FDG uptake in the ROI containing the thymic lesion is 4.0. MDCT image showed apparently well-defined thymic mass, indicating no adjacent organs infiltration; however, pathology revealed thymic capsula microinfiltration, with microscopic tumor spread into mediastinal fat.

different and the nature of the lesions might have been described as potentially benign. Although the false positive cases reduce PET specificity to 87.5%, it does not influence the high sensitivity (100%). Several studies reported that semi-quantitative analysis is not reliable to differentiate thymic lesions nature, while recently Sung et al. [23] reported that SUV is correlated with malignant grade of thymic lesions. In a group of 33 patients with 8 low-risk thymoma, 9 high-risk thymoma and 16 thymic carcinoma, they showed that SUV is, respectively, 4 ± 0.42, 5.6 ± 1.90 and 10.5 ± 4.68 and reported a statistically significant difference between thymic carcinoma and the other thymic lesions. Our cohort of patients included some bearing benign lesions and we were able to demonstrate that SUVs do differentiate benign lesions from malignant ones (p = 0.003) (Fig. 3). Therefore, a negative [18 F]FDG-PET/CT should suggest that lesion is not to be treated with an aggressive treatment, such as surgery, because it is likely to be benign. Among malignant lesions our results showed no statistical differences between low and high-risk group (p = 0.37) so PET does not seem to enable stratification for malignant grade. Our only two thymic carcinoma patients showed a SUVbw max of 8 and 9.5 that is in line with the results reported by Sung et al. [23] in the thymic carcinoma group (10.5 ± 4.68) and by Brink et al. [1] that reported in a thymic carcinoma a SUV of 9.6. Obviously we cannot perform a statistical analy-

sis on two patients but we observe an increasing trend of SUV similarly to Sung. The low number of patients could prevent the test to reach statistical significance. Noteworthy, [18 F]FDG-PET/CT detected an unknown neoplastic lesions in a patient, a Warthin lesion.

5. Conclusions In the cases reported hereby, neither imaging modality by itself was sufficient to identify the nature of thymic lesions. Both MDCT and [18 F]FDG-PET/CT are needed in thymic lesion nature assessment and treatment planning. In particular lesions characterized by negative [18 F]FDG-PET/CT with SUV below 2.2 and no visible infiltration of mediastinum fat or adjacent organs at MDCT are so likely to be benign that an invasive approach should not be offered as first line procedure.

Conflict of interest None declared.

References [1] Brink I, Reinhardt MJ, Hoegerle S, Altehoefer C, Moser E, Nitzsche EU. Increased metabolic activity in the thymus gland studied with 18F-FDG PET: age dependency

368

[2]

[3]

[4]

[5] [6]

[7]

[8] [9]

[10]

[11]

[12] [13]

[14]

L.L. Travaini et al. and frequency after chemotherapy. J Nucl Med 2001;42: 591—5. Dustin ML, Rothlein R, Bhan AK, Dinarello CA, Springer TA. Induction by IL 1 and interferon-gamma: tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-1). J Immunol 1986;137:245—54. Tatebe S, Oka K, Uehara A, Shinonaga M, Kuraoka S. Unusual remnant thymic tissue in an adult mimicking malignant neoplasm: escape from age-related involution. Thorac Cardiovasc Surg 2006;54:138—40. Choi YW, McAdams HP, Jeon SC, Hong EK, Kim YH, Im JG, et al. Idiopathic multilocular thymic cyst: CT features with clinical and histopathologic correlation. Am J Roentgenol 2001;177:881—5. Rosai J. Histological Typing of tumours of the thymus. 2nd ed. New York, NY: Springer-Verlag; 1999. Osserman KE, Genkins G. Studies in myasthenia gravis: review of a 20-year experience in over 1200 patients. Mt Sinai J Med 1971;38:497—537. Ogawa K, Toita T, Uno T, Fuwa N, Kakinohana Y, Kamata M, et al. Treatment and prognosis of thymic carcinoma: a retrospective analysis of 40 cases. Cancer 2002;94:3115—9. Wick MR, Scheithauer BW, Weiland LH, Bernatz PE. Primary thymic carcinomas. Am J Surg Pathol 1982;6:613—30. Travis W, Brambilla E, M¨ uller-Hermelink H. Pathology and genetics of tumours of the lung, pleura, thymus and heart. Paris, France: Oxford University Press; 2004. Schmidt-Wolf IG, Rockstroh JK, Sch¨ uller H, Hirner A, Grohe C, M¨ uller-Hermelink HK, et al. Malignant thymoma: current status of classification and multimodality treatment. Ann Hematol 2003;82:69—76. Wittram C, Fischman AJ, Mark E, Ko J, Shepard JA. Thymic enlargement and FDG uptake in three patients: CT and FDG positron emission tomography correlated with pathology. AJR Am J Roentgenol 2003;180:519—22. Maher MM, Shepard JA. Imaging of thymoma. Semin Thorac Cardiovasc Surg 2005;17:12—9. Dwamena BA, Sonnad SS, Angobaldo JO, Wahl RL. Metastases from non—small cell lung cancer: mediastinal staging in the 1990s—–meta-analytic comparison of PET and CT. Radiology 1999;213:530—6. Gearhart SL, Frassica D, Rosen R, Choti M, Schulick R, Wahl R. Improved staging with pretreatment positron emission tomography/computed tomography in low rectal cancer. Ann Surg Oncol 2006;13:397—404.

[15] Cerfolio RJ, Bryant AS. Maximum standardized uptake values on positron emission tomography of esophageal cancer predicts stage, tumor biology, and survival. Ann Thorac Surg 2006;82:391—5. [16] Fogarty GB, Peters LJ, Stewart J, Scott C, Rischin D, Hicks RJ. The usefulness of fluorine 18-labelled deoxyglucose positron emission tomography in the investigation of patients with cervical lymphadenopathy from an unknown primary tumor. Head Neck 2003;25:138—45. [17] Kubota K, Yamada S, Kondo T, Yamada K, Fukuda H, Fujiwara T, et al. PET imaging of primary mediastinal tumours. Br J Cancer 1996;73:882—6. [18] Ferdinand B, Gupta P, Kramer EL. Spectrum of thymic uptake at 18F-FDG PET. RadioGraphics 2004;24:1611—6. [19] Sasaki M, Kuwabara Y, Ichiya Y, Ichiya Y, Akashi Y, Yoshida T, et al. Differential diagnosis of thymic tumors using a combination of 11C-methionine PET and FDG PET. J Nucl Med 1999;40:1595—601. [20] Nakahara T, Fujii H, Ide M, Nishiumi N, Takahashi W, Yasuda S, et al. FDG uptake in the morphologically normal thymus: comparison of FDG positron emission tomography and CT. Br J Radiol 2001;74:821—4. [21] Castellucci P, Nanni C, Farsad M, Alinari L, Zinzani P, Stefoni V, et al. Potential pitfalls of 18F-FDG PET in a large series of patients treated for malignant lymphoma: prevalence and scan interpretation. Nucl Med Commun 2005;26:689—94. [22] Masaoka A, Monden Y, Nakahara K, Tanioka T. Follow-up study of thymomas with special reference to their clinical stages. Cancer 1981;48:2485—92. [23] Sung YM, Lee KS, Kim BT, Choi JY, Shim YM, Yi CA. 18F-FDG PET/CT of thymic epithelial tumors: usefulness for distinguishing and staging tumor subgroups. J Nucl Med 2006;47:1628—34. [24] Hamacher K, Coenen HH, Stocklin G. Efficient stereospecific synthesis of no-carrier-added 2-[18 F]-fluoro-2-deoxy-D-glucose using aminopolyether supported nucleophilic substitution. J Nucl Med 1986;27:235—8. [25] Jung KJ, Lee KS, Han J, Kim J, Kim TS, Kim EA. Malignant thymic epithelial tumors: CT-pathologic correlation. AJR Am J Roentgenol 2001;176:433—9. [26] Restrepo CS, Pandit M, Rojas IC, Villamil MA, Gordillo H, Lemos D, et al. Imaging findings of expansile lesions of the thymus. Curr Probl Diagn Radiol 2005;34:22—34. [27] Halaweh O, Usmani SZ, Bloom R. Chemotherapy-induced thymus hyperplasia can mimic tumor recurrence. Breast J 2006;12:381—2.