Eur J Nucl Med Mol Imaging (2013) 40:10361043 DOI 10.

1007/s00259-013-2388-9

ORIGINAL ARTICLE

Diagnostic accuracy of 18F-FDG PET/CT for detecting recurrence in patients with primary skeletal Ewing sarcoma
Punit Sharma & Bangkim Chandra Khangembam & K. C. Sudhir Suman & Harmandeep Singh & Sishir Rastogi & Shah Alam Khan & Sameer Bakhshi & Sanjay Thulkar & Chandrasekhar Bal & Arun Malhotra & Rakesh Kumar

Received: 20 December 2012 / Accepted: 4 March 2013 / Published online: 5 April 2013 # Springer-Verlag Berlin Heidelberg 2013

Abstract Purpose To evaluate the diagnostic accuracy of 18F-FDG PET/CT for detecting recurrence in patients with primary skeletal Ewing sarcoma. Methods We retrospectively analysed data from 53 patients (age 20.110.5 years, 39 male) who had undergone 71 18FFDG PET/CT studies for suspected recurrence (52 studies) or for routine follow-up (19 studies) after primary therapy of skeletal Ewing sarcoma. 18F-FDG PET/CT studies were evaluated qualitatively and quantitatively (maximum standardized uptake value, SUVmax) by two nuclear medicine physicians in consensus. Sensitivity, specificity, predictive values and accuracy were calculated on per study basis. Clinical/imaging follow-up (minimum 6 months) and/or histopathology (when available) were taken as the reference standard. Results Of the total of 71 18F-FDG PET/CT studies, 42 (59.1 %) were positive for recurrence and 29 (40.9 %) were
P. Sharma : B. C. Khangembam : K. C. S. Suman : H. Singh : C. Bal : A. Malhotra : R. Kumar (*) Department of Nuclear Medicine, All India Institute of Medical Sciences, E-81, Ansari Nagar (East), AIIMS Campus, New Delhi 110029, India e-mail: rkphulia@hotmail.com S. Rastogi : S. A. Khan Department of Orthopaedics, All India Institute of Medical Sciences, New Delhi, India S. Bakhshi Department of Medical Oncology, Institute Rotary Cancer Hospital, All India Institute of Medical Sciences, New Delhi, India S. Thulkar Department of Radiodiagnosis, All India Institute of Medical Sciences, New Delhi, India

negative for recurrence. Local recurrence was most common (38 studies) followed by bone metastasis (9 studies), and node and lung metastasis (2 studies each). Of the 71 studies, 38 were true-positive, 27 were true-negative, 4 were falsepositive and 2 were false-negative. Overall per study based sensitivity was 95 %, specificity was 87 %, PPV was 90 %, NPV was 93 % and accuracy was 91.5 %. No significant difference was found in the accuracy of PET/CT between the suspected recurrence group and the routine follow-up group (94 % vs. 84 %; P =0.390). Overall mean lesion SUVmax was 7.84.1 (range 1.917.2). No site-based difference was found in SUVmax. Conclusion 18F-FDG PET/CT demonstrates high diagnostic accuracy for detecting recurrence in patients with primary skeletal Ewing sarcoma, when it is suspected (clinically or on imaging) or during routine follow-up. Keywords Ewing sarcoma . Skeleton . Recurrence . F-FDG . PET/CT . Accuracy

18

Introduction Ewing sarcomas are highly anaplastic small, round, blue cell tumours, thought to be of neuroectodermal origin [1]. They are the second most common primary bone tumour [2]. These tumours occur more frequently in males than in females, and are associated with increased incidence in white and Hispanic children [1]. About 75 % of Ewing sarcoma arise from bone and the remainder from soft tissue, the latter being more common in adult patients. With the use of multimodal therapy in the form of chemotherapy and surgery with or without radiotherapy, survival has improved to 6070 % [3]. Unfortunately, 3040 % patients will

Eur J Nucl Med Mol Imaging (2013) 40:10361043

1037

develop local and/or distant recurrent disease typically between 2 and 10 years after diagnosis [4]. Survival following relapse is 20 %, and despite the introduction of new therapeutic agents including high-dose chemotherapy, attempts to improve the prognosis have not been very successful [5]. Detection of recurrent disease can have a significant impact on patient management and prognosis. Imaging plays a pivotal role in the diagnosis of recurrence. Most commonly radiography, CT, MRI and bone scan (BS) are employed [6]. However, posttherapy changes reduce the specificity of anatomical imaging. In addition, many of these patients have metallic implants which can introduce significant artefacts, further compromising the usefulness of these imaging approaches. Given its low specificity, BS has a limited role for detecting recurrence. 18F-FDG PET/CT combines the advantages of metabolic imaging and anatomical imaging in a single setting. A recent metaanalysis by Treglia et al. [7] has demonstrated the utility of 18F-FDG PET and PET/CT in the evaluation of Ewing sarcoma family of tumours. While the role of 18F-FDG PET/CT in staging of Ewing sarcoma is well established [7], its role in detecting recurrence in such tumours is less clear. The few studies that have evaluated the role of 18 F-FDG PET or PET/CT in recurrent Ewing sarcoma suffer from certain drawbacks including mixed histopathological populations (39 and 22 patients with Ewing sarcoma [8, 9]), variable indications for PET/CT (53 patients [10]), often along with a small sample size (14, 9 and 11 patients [1113]). In addition many of these studies have employed PET alone [8, 13], which is inferior to PET/CT for the evaluation of Ewing sarcoma tumours [10]. None of the studies evaluated 18F-FDG PET/CT in patients with primary skeletal Ewing sarcoma alone. Hence, the aim of the present study was to evaluate the diagnostic accuracy of 18F-FDG PET/CT for detecting recurrence in a large population of patients with primary skeletal Ewing sarcoma.

Institutional ethics committee clearance was obtained. Because of the retrospective nature of the study written informed consent was waived.
18

F-FDG PET/CT acquisition

The patients fasted for 46 h before PET/CT. Blood glucose was less than 140 mg/dl in all patients. 18F-FDG at a dose of 370 MBq (10 mCi) in adults and 67 MBq/kg (0.160. 18 mCi/kg) in children was injected intravenously. The patients rested in a quiet room and after 4560 min for 18F-FDG uptake, PET/CT imaging was done. All studies were acquired on a dedicated PET/CT scanner (Biograph 2; Siemens, Erlangen, Germany). No intravenous contrast agent was administered for the CT part of the PET/CT scan. CT acquisition was performed with a spiral dual slice CT scanner with 130 kV, 60 mAs, 4 mm slice thickness and a pitch of 1. Images were acquired using a matrix of 512512 pixels and a pixel size of 1 mm. After CT acquisition, the PET acquisition of the same axial range was done with the patient in the same position. The threedimensional PET acquisition was done with 23 min per bed position. PET data were acquired using a matrix of 128128 pixels with a slice thickness of 1.5 mm. CT-based attenuation correction of the emission images was performed. PET images were reconstructed using the iterative ordered subsets expectation maximization method (two iterations and eight subsets). After completion of the PET acquisition, the reconstructed attenuation-corrected PET images, CT images and fused matched PET and CT images were available for review in the axial, coronal and sagittal planes, as well as maximum intensity projections (MIP) and in three-dimensional cine mode.
18

F-FDG PET/CT image analysis

Materials and methods Data from patients with skeletal Ewing sarcoma who had undergone 18F-FDG PET/CT between March 2006 and January 2012 were retrospectively analysed from the departmental registry. Patients were included if they had histopathologically proven skeletal Ewing sarcoma, had undergone previous treatment, had undergone 18F-FDG PET/CT for suspected recurrence (clinical/imaging) or for routine follow-up, and had a minimum of 6 months of clinical/imaging follow up after PET/CT. Patients were excluded if they had extraskeletal Ewing sarcoma, had not undergone previous treatment, and/or had undergone PET/CT before <4 weeks of treatment, and/or had less than 6 months clinical/imaging follow-up after PET/CT. A total of 53 patients who had undergone 71 18FFDG PET/CT studies were included in the final analysis.

F-FDG PET/CT studies were retrospectively evaluated by two experienced nuclear medicine physicians with more than 5 years experience in oncological PET/CT imaging (R.K. 10 years. P.S. 5 years). Both of them were unaware of other clinical or imaging information at the time of review except for the primary diagnosis, and interpreted the 18F-FDG PET/CT images in consensus. Any area of increased 18F-FDG uptake was looked for, keeping in mind the physiological tracer distribution. In patients with metallic implants or a prosthesis in situ, images without attenuation correction were also evaluated to rule out overcorrection artefacts. Positive findings on 18F-FDG PET were localized to anatomical images from the nonenhanced CT scan. However, pulmonary lesions showing the characteristic appearance of metastasis on CT scan (well-circumscribed rounded lesions with soft tissue attenuation, location in the periphery of the lung, absence of calcification, multiple nodules of variable size, canon ball opacities, presence of feeding vessel sign) were taken as positive even in the

18

1038

Eur J Nucl Med Mol Imaging (2013) 40:10361043

absence of significant 18F-FDG uptake. PET/CT images were interpreted as positive or negative for recurrence. PET/CT findings were divided into five major groups: local recurrence, nodal metastasis, pulmonary metastasis, skeletal metastasis and other sites. Semiquantitative analysis was also done using the standardized uptake value (SUVmax). To minimize the effects of tumoral metabolic inhomogeneities, the region of interest (ROI) placed over the tumour was as large as possible to encompass the entire cross-sectional area of the lesion. For each lesion, the SUVmax was measured twice and averaged to yield the final value. The SUV was calculated using the default method in terms of body weight: (SUV = mean ROI activity (MBq/g)/injected dose (MBq)/body weight (g). The lesion with the maximum SUVmax in each group was taken for analysis while the highest SUVmax was taken for overall analysis. Reference standard A combination of clinical follow-up and imaging follow-up (CT, MRI, PET/CT; minimum 6 months, median 16 months, range 649 months) and/or histopathology (when available) was taken as the reference standard. Lesions that had increased in size on radiological follow-up and/or showed an increase in 18 F-FDG uptake on follow-up PET/CT were taken as truepositive. Similarly, lesions showing response to therapy were taken as true-positive. Lesions not showing any change or those showing a reduction in size or 18F-FDG uptake on follow-up without any therapy were taken as false-positive. Results of the PET/CT scan were analysed on a per-study basis. Statistical analysis All the data analyses were performed using the statistical software packages SPSS 11.5 (SPSS Inc., Chicago, IL). Categorical data are expressed as number and percentage, while continuous data are expressed as mean and standard deviation. Sensitivity, specificity, accuracy, positive and negative predictive value (PPV, NPV) of 18F-FDG PET/CT were calculated on a per-study basis with 95 % confidence intervals. The Mann-Whitney test with two-tailed probability was used to compare continuous variables. The chi-squared (2) test was used to compare the diagnostic accuracy of 18F-FDG PET/CT between the suspected relapse group and the routine follow-up group. P values <0.05 were considered significant.

study. Among these 71 18F-FDG PET/CT studies, 52 were done for suspected recurrence (based on clinical findings in 32 and imaging findings in 20) while 19 were done for routine follow-up. Patient characteristics including age, sex, primary histology and indication for PET/CT are detailed in Table 1. Of the 53 patients, 48 (90.64 %) were aged less than 30 years and 37 (69.8 %) were aged less than 20 years. The primary site was the appendicular skeleton in 30 patients (humerus in 2, hand in 1, femur in 10, tibia in 7, fibula in 7, and foot in 3) and the axial skeleton in 23 (skull and calvaria in 4, spine in 4, pelvis in 12, clavicle in 1, rib in 1, and scapula in 1). Reference standard Based on the reference standard detailed in Materials and methods, 56.3 % of patients (40/71) were positive for relapse and 43.7 % (31/71) were negative for relapse at the time of the PET/CT scan. Histopathology was available only for 27 PET/CT studies (39 lesions) as histopathological confirmation of all lesions was not technically or ethically feasible. Bone marrow biopsy was done in two patients. The final diagnosis was established with clinical/imaging follow-up for 44 PET/CT studies. The follow-up imaging was contrast-enhanced CT (in 28 patients) and/or MRI (in 17) and/or PET/CT (in 21). Five patients died of progressive disease during follow-up, while seven patients showed clinically apparent recurrent disease.
18

F-FDG PET/CT imaging findings

Of the total of 71 18F-FDG PET/CT studies, 42 (59.1 %) were positive for recurrence and 29 (40.9 %) were negative for recurrence, and 38 were true-positive (Figs. 1, 2 and 3), 27 were true-negative, 4 were false-positive and 2 were false-negative. The overall per-study sensitivity, specificity,

Table 1 Patient characteristics Characteristic Total no. of patients Age (years) Mean SD Median (Range) Sex, n (%) Male Female Total no. of 18F-FDG PET/CT studies Indication for 18F-FDG PET/CT, n (%) Suspected recurrence Routine follow-up Value 53 20.110.5 18 (758) 39 (73.5) 14 (26.5) 71 52 (73.2) 19 (26.8)

Results Patient characteristics A total 53 patients with skeletal Ewing sarcoma who had undergone 7118F-FDG PET/CT studies were included in this

Eur J Nucl Med Mol Imaging (2013) 40:10361043 Fig. 1 A 30-year-old man with Ewing sarcoma of the left proximal tibia, 7 months after chemotherapy. 18F-FDG PET/ CT was done for suspected recurrence. af PET (a, c, e) and PET/CT (b, d, f) images of the tibia in the transaxial (a, b), coronal (c, d) and sagittal (e, f) planes reveal a lytic/sclerotic lesion of the left proximal tibia (arrows) with intense 18F-FDG uptake (SUVmax 9.3). g The whole-body MIP PET image is unremarkable. Local recurrence was diagnosed by 18F-FDG PET/CT and was confirmed after secondary surgery

1039

PPV, NPV and accuracy of 18F-FDG PET/CT are presented in Table 2. Of the four false-positive PET/CT studies, three were false-positive for local recurrence (two confirmed on follow-up imaging and one on histopathology). One was false-positive for mediastinal nodal metastasis (confirmed as tuberculosis by response to antitubercular drugs). Both false-negative studies were due to local recurrence (one confirmed on histopathology, the other showing an increase in size and 18F-FDG uptake on follow-up PET/CT). In the suspected recurrence group (52 studies), 35 were true-positive, 14
Fig. 2 A 19-year-old man with Ewing sarcoma of the right chest wall, after surgery and chemotherapy. 18F-FDG PET/ CT was done for suspected recurrence based on CT findings which showed a right 6th rib soft tissue mass. a Whole-body MIP PET image shows irregular 18F-FDG uptake in the right lower chest wall (arrow). b, c Transaxial PET (b) and PET/CT (c) images of the thorax show a soft-tissue mass arising from the right sixth rib (arrows) with mild and heterogeneous 18F-FDG uptake (SUVmax 3.1). The findings are suggestive of recurrent disease which was confirmed on follow-up imaging

were true-negative, 1 was false-positive and 2 were false-negative. In the routine follow-up group (19 studies), 3 PET/CT were true-positive, 13 were truenegative and 3 were false-positive. None of the PET/CT studies was false-negative in the follow up group. The per-study sensitivity, specificity, PPV, NPV and accuracy of 18F-FDG PET/CT in the suspected recurrence and follow-up groups are presented in Table 2. No significant difference was found in the accuracy of PET/CT between the suspected recurrence and routine follow-up group (P = 0.390).

1040 Fig. 3 A 26-year-old man with Ewing sarcoma of the left fibula, after surgery and chemotherapy. Eight months after completing therapy the patient presented with left upper leg swelling. He underwent 18FFDG PET/CT. a, b PET (a) and PET/CT (b) images of the left leg show an ill-defined softtissue mass with involvement of the left tibia and adjacent muscles (bold arrow), showing increased 18F-FDG uptake (SUVmax 15.6). c Whole-body MIP PET image shows multiple focal areas of 18F-FDG uptake throughout the skeleton including the bones of the left leg (a, arrows), suggesting bone marrow metastasis. The patient was started on chemotherapy, but died of progressive disease

Eur J Nucl Med Mol Imaging (2013) 40:10361043

The sites of recurrence as seen on PET/CT are detailed in Table 3 (Figs. 1, 2, 3 and 4). Local recurrence was most common (38 studies). The next most common site was bone (n =8) followed by lymph node and lung (n =7). The lymph node groups involved most commonly on PET/CT were mediastinal (n =8), while popliteal and common iliac nodal metastasis was seen in 1 each. Other uncommon sites of metastasis included bone marrow (n =2), muscle (n =2) and subcutaneous soft tissue (n =2). Of the 68 lesion groups demonstrated on PET/CT (Table 3), 25 were not clinically obvious and/or diagnosed on conventional imaging. These included 11 local recurrences, 6 nodal lesions (all mediastinal), 4 bone metastasis, 2 bone marrow metastases, 1 muscle and 1 soft tissue metastasis. Semiquantitative analysis The lesion with the highest SUVmax on each PET/CT study was used for semiquantitative analysis. On semiquantitative analysis the overall mean lesion SUVmax was 7.84.1 (median 7.9, range 1.917.2). The mean SUVmax of local lesions
Table 2 Per-study diagnostic accuracy of 18F-FDG PET/CT Parameter Sensitivity (%) Specificity (%) PPV (%) NPV (%) Accuracy (%) Overall 95 (95 87 (95 90 (95 93 (95 91.5 % CI % CI % CI % CI

was 7.94.2 (median 7.4, range 1.917.2), of nodal lesions was 6.12 (median 6, range 3.89.1), and of bone lesions was 6.61.7 (median 7, range 2.88.6). No significant differences were found between the SUVmax of local lesions and those of nodal lesions (P =0.486) and bone lesions (P =0.598), or between the SUVmax of nodal lesions and that of bone lesions (P =0.630). As many pulmonary lesions were small and did not show significant 18F-FDG uptake, and for these lesions semiquantitative analysis was not done.

Discussion F-FDG PET/CT images the glucose metabolism in Ewing sarcoma and 18F-FDG uptake correlates with the level of glucose transporter 1 expression in these tumours [14]. It has been shown to be very useful for grading, staging and monitoring response to therapy of primary Ewing sarcoma [1517]. However, the role of 18F-FDG PET/CT in detecting recurrence of such tumours is less well defined [7]. In contrast to primary Ewing sarcoma, patients with recurrence
Suspected recurrence group 8399) 7096) 7797) 7799) 95 (95 99 (95 97 (95 87.5 (95 94 % CI % CI % CI % CI 8299) 68100) 85100) 6298) Follow-up group 100 (95 81 (95 50 (95 100 (95 84 % % % % CI CI CI CI 29100) 54100) 1288) 75100)
18

Eur J Nucl Med Mol Imaging (2013) 40:10361043 Table 3 Sites of recurrence on Site
a 18

1041

F-FDG PET/CT No. (%) of studies 38 (90.4) 8 (19) 8 (19) 9 (21.4) 5 (11.9)
b

Local Lymph node Lung Bone Others

a b

Some patients had more than one site of recurrence on 18 F-FDG PET/CT. Of 42 positive
18

F-FDG PET/CT studies.

have a poor prognosis and the timing and type of recurrence are important prognostic factors in these patients [18, 19]. With the recent advances in therapy of recurrent Ewing sarcoma [20], detection of recurrence and localization of the site are important for both management and prognosis. Franzius et al. [8] and Mody et al. [13] evaluated 18F-FDG PET alone for detection of recurrence in small, mixed populations of malignant primary bone tumours including Ewing sarcoma, and found it to be useful. Both the studies used PET alone, which because of its limited spatial resolution cannot show the precise anatomical location of identified lesions and is inferior to PET/CT in Ewing sarcoma [10]. Very few studies have evaluated the diagnostic accuracy of 18F-FDG PET/CT in recurrent Ewing sarcoma, and most evaluated small or mixed patient populations. Gerth et al. [10] evaluated 163 PET/CT studies in 53 patients with Ewing tumour, including 68 PET/CT scans done for restaging and follow-up. They demonstrated that PET/CT is more accurate than PET alone (P <0.001) with a sensitivity of 87 % and specificity of 97 %. However, as the diagnostic values of 18F-FDG PET/CT in the recurrence

group were not mentioned separately, a clear conclusion regarding its utility for recurrent Ewing tumour could not be drawn. Arush et al. [12] also demonstrated the utility of 18 F-FDG PET/CT for detecting local recurrence and distant metastasis in a small mixed paediatric population of 19 patients with sarcoma (Ewing sarcoma 9). More recently, Charest et al. [9] retrospectively analysed the usefulness of 18 F-FDG PET/CT in 212 patients with bone and soft tissue sarcoma, including 22 patients with Ewing sarcoma. However, only 3 of these 22 patients underwent PET/CT for suspected recurrence. Given these limitations of the available literature and lack of systematic studies, we aimed to evaluate the diagnostic accuracy of 18 F-FDG PET/CT for detecting recurrence in primary skeletal Ewing sarcoma. 18 F-FDG PET/CT showed a high sensitivity of 95 %, a specificity of 87 % and an accuracy of 91.5 % for detecting recurrent skeletal Ewing sarcoma. These results are consistent with values reported previously [9, 10, 12]. In patients with suspected recurrence, 18F-FDG PET/CT showed high sensitivity (95 %) and specificity (99 %). Thus 18F-FDG PET/CT can be used to confirm or rule out recurrence in patients with clinical/imaging suspicion. Given the limited specificity of conventional imaging (CT, MRI, BS) in differentiating posttherapy changes from recurrence [6], 18F-FDG PET/CT will be very useful. We found that 18F-FDG PET/CT was able to accurately detect recurrence during routine follow-up (sensitivity 100 %, specificity 81 %) and therefore when available could be integrated into management. The accuracy of 18FFDG PET/CT was not significantly different between those with suspected recurrence and those undergoing routine follow-up (94 % vs. 84 %, P =0.390). This finding was similar to that reported by Charest et al. [9].

Fig. 4 A 15-year-old boy with Ewing sarcoma of the right tibia, 8 months after surgery and bone grafting. He underwent 18F-FDG PET/CT for routine follow-up. a PET image of the right leg shows focal 18F-FDG uptake (SUVmax 6.1) in the right tibia (arrow). b, c PET (b) and PET/CT (c) images localize this focus to the bone grafting site (arrows), suspicious

for local recurrence. d, e A solitary pulmonary nodule can also be seen in the left lung lower lobe periphery on the CT image (d, arrow) with mild 18 F-FDG uptake on the PET/CT image (e, arrow), suggesting pulmonary metastasis. Both the local lesion and the pulmonary nodule disappeared after chemotherapy, confirming the diagnosis

1042

Eur J Nucl Med Mol Imaging (2013) 40:10361043

Local recurrence was the most common recurrence (90. 4 % of positive PET/CT studies). The presence of a prosthesis, asymmetrical weigh bearing, posttherapy changes, and the especially difficult problem of attenuation overcorrection because of metallic implants/prostheses or posttherapy callous formation make interpretation of PET/CT studies difficult in these patients [2, 21]. Hence, images without attenuation correction should always be evaluated when PET/CT is positive for local recurrence. In this study re-evaluation of images without attenuation correction changed the PET/CT finding of five studies from positive to negative, all of which were true-negative for local recurrence. The use of newer metal artefact reduction strategies might further reduce this shortcoming [22]. Bone was the second most common site of recurrence on PET/CT studies (21.4 %), followed by the lung and lymph nodes. When evaluating lung lesions, we relied more on CT images and less on 18F-FDG uptake [23]. If the patient had pulmonary nodules with morphological features of metastasis, they were taken as positive irrespective of 18F-FDG uptake as underestimation of 18F-FDG uptake is seen in small pulmonary nodules due to the partial volume effect [24]. Unfortunately, CT alone is unable to distinguish with certainty between benign and metastatic pulmonary nodules in children with known malignant solid tumours [25]. Interestingly, a recent study has demonstrated that SUVmax of >1. 09 or SUVratio (SUVmax nodule/SUVmax mediastinum) of >0.83 can differentiate between benign and metastatic pulmonary nodules seen on CT in paediatric sarcomas [26]. In the present study of the 68 lesion groups seen on PET/CT, 26 were unexpected, i.e. not clinically obvious/diagnosed on conventional imaging. As 18F-FDG uptake of bone sarcomas depends on histological grade [14], recurrent tumours, which are usually more aggressive, will show higher 18F-FDG uptake and SUVmax. On semiquantitative analysis, the SUVmax of recurrent lesions was high (median SUVmax 7.4 for local recurrence, 6 for nodal disease, and 7 for bone lesions). However, there was wide variation in SUVmax (range 1. 917.2), and hence no definite cut-off can be recommended. No significant differences was seen in SUVmax among local, nodal and bone lesions. Because of low 18F-FDG uptake in many of pulmonary lesions, their semiquantitative analysis was not attempted. The limitations of the present study include its retrospective design which might have introduced a selection bias. The lack of a histopathological reference standard in all patients was a major limitation, possibly biasing the specificity of the results. Although, this would have been ideal, in many patients it was deemed unethical because of the presence of metastases at distant sites on imaging or was not attempted for technical reasons. In addition, we were not able to compare the results of 18F-FDG PET/CT directly

with those of conventional imaging (CT, MRI, BS) because of lack of data, which might have led to a falsely high sensitivity for PET/CT. To minimize these potential biases, imaging/clinical follow-up of at least 6 months was required. Larger prospective multicentre studies, including direct comparison with conventional imaging, might overcome the shortcomings of the present study. Conclusion The results of this retrospective study demonstrate the high diagnostic accuracy of 18 F-FDG PET/CT for detecting recurrence in patients with primary skeletal Ewing sarcoma. 18F-FDG PET/CT is highly accurate in patients in whom recurrence is suspected clinically or on imaging, as well as in patients undergoing PET/CT for routine follow-up.
No funding received from any organisation for this study. Conflicts of interest None.

References
1. Heare T, Hensley MA, DellOrfano S. Bone tumors: osteosarcoma and Ewings sarcoma. Curr Opin Pediatr. 2009;21:36572. 2. Hawkins DE, Blling T, DuBois SG, et al. Ewing sarcoma. In: Pizzo PA, Poplack DG, editors. Principles and practice of pediatric oncology. Philadelphia: Lippincott Williams & Wilkins; 2010. p. 9871014. 3. Esiashvili N, Goodman M, Marcus Jr RB. Changes in incidence and survival of Ewing sarcoma patients over the past 3 decades: Surveillance Epidemiology and End Results data. J Pediatr Hematol Oncol. 2008;30:42530. 4. Wasilewski-Masker K, Liu Q, Yasui Y, Leisenring W, Meacham LR, Hammond S, et al. Late recurrence in pediatric cancer: a report from the Childhood Cancer Survivor Study. J Natl Cancer Inst. 2009;101:170920. 5. Rodriguez-Galindo C, Billups CA, Kun LE, Rao BN, Pratt CB, Merchant TE, et al. Survival after recurrence of Ewing tumors: the St. Jude Childrens Research Hospital experience, 19791999. Cancer. 2002;94:5619. 6. Reuther G, Mutschler W. Detection of local recurrent disease in musculoskeletal tumors: magnetic resonance imaging versus computed tomography. Skeletal Radiol. 1990;19:8590. 7. Treglia G, Salsano M, Stefanelli A, Mattoli MV, Giordano A, Bonomo L. Diagnostic accuracy of 18F-FDG-PET and PET/CT in patients with Ewing sarcoma family tumours: a systematic review and a meta-analysis. Skeletal Radiol. 2012;41:24956. 8. Franzius C, Daldrup-Link HE, Wagner-Bohn A, Sciuk J, Heindel WL, Jrgens H, et al. FDG-PET for detection of recurrences from malignant primary bone tumors: comparison with conventional imaging. Ann Oncol. 2002;13:15760. 9. Charest M, Hickeson M, Lisbona R, Novales-Diaz JA, Derbekyan V, Turcotte RE. FDG PET/CT imaging in primary osseous and soft tissue sarcomas: a retrospective review of 212 cases. Eur J Nucl Med Mol Imaging. 2009;36:194451. 10. Gerth HU, Juergens KU, Dirksen U, Gerss J, Schober O, Franzius C. Significant benefit of multimodal imaging: PET/CT compared

Eur J Nucl Med Mol Imaging (2013) 40:10361043 with PET alone in staging and follow-up of patients with Ewing tumors. J Nucl Med. 2007;48:19329. McCarville MB, Christie R, Daw NC, Spunt SL, Kaste SC. PET/CT in the evaluation of childhood sarcomas. AJR Am J Roentgenol. 2005;184:1293304. Arush MW, Israel O, Postovsky S, Militianu D, Meller I, Zaidman I, et al. Positron emission tomography/computed tomography with 18fluoro-deoxyglucose in the detection of local recurrence and distant metastases of pediatric sarcoma. Pediatr Blood Cancer. 2007;49:9015. Mody RJ, Bui C, Hutchinson RJ, Yanik GA, Castle VP, Frey KA, et al. FDG PET imaging of childhood sarcomas. Pediatr Blood Cancer. 2010;54:2227. Tateishi U, Yamaguchi U, Seki K, Terauchi T, Arai Y, Hasegawa T. Glut-1 expression and enhanced glucose metabolism are associated with tumour grade in bone and soft tissue sarcomas: a prospective evaluation by [18F]fluorodeoxyglucose positron emission tomography. Eur J Nucl Med Mol Imaging. 2006;33:68391. Bastiaannet E, Groen H, Jager PL, Cobben DC, van der Graaf WT, Vaalburg W, et al. The value of FDG-PET in the detection, grading and response to therapy of soft tissue and bone sarcomas; a systematic review and meta-analysis. Cancer Treat Rev. 2004;30:83101. Vlker T, Denecke T, Steffen I, Misch D, Schnberger S, Plotkin M, et al. Positron emission tomography for staging of pediatric sarcoma patients: results of a prospective multicenter trial. J Clin Oncol. 2007;25:543541. Denecke T, Hundsdrfer P, Misch D, Steffen IG, Schnberger S, Furth C, et al. Assessment of histological response of paediatric bone sarcomas using FDG PET in comparison to morphological volume measurement and standardized MRI parameters. Eur J Nucl Med Mol Imaging. 2010;37:184253.

1043 18. Leavey PJ, Mascarenhas L, Marina N, Chen Z, Krailo M, Miser J, et al. Prognostic factors for patients with Ewing sarcoma (ES) at first recurrence following multimodal therapy: a report from Children s Oncology Group. Pediatr Blood Cancer. 2008;51:3348. 19. Robinson SI, Ahmed SK, Laack NN, Rose PS, Okuno SH. Clinical outcomes of adult patients with relapsed Ewing sarcoma: a 30-year single institution experience. J Clin Oncol. 2011;29(Suppl):10041. 20. Barker LM, Pendergrass TW, Sanders JE, Hawkins DS. Survival after recurrence of Ewings sarcoma family of tumors. J Clin Oncol. 2005;23:435462. 21. Blodgett TM, Mehta AS, Mehta AS, Laymon CM, Carney J, Townsend DW. PET/CT artifacts. Clin Imaging. 2011;35:4963. 22. Abdoli M, Dierckx RA, Zaidi H. Metal artifact reduction strategies for improved attenuation correction in hybrid PET/CT imaging. Med Phys. 2012;39:334360. 23. Franzius C, Daldrup-Link HE, Sciuk J, Rummeny EJ, Bielack S, Jrgens H, et al. FDG-PET for detection of pulmonary metastases from malignant primary bone tumors: comparison with spiral CT. Ann Oncol. 2001;12:47986. 24. Chang G, Chang T, Pan T, Clark Jr JW, Mawlawi OR. Joint correction of respiratory motion artifact and partial volume effect in lung/thoracic PET/CT imaging. Med Phys. 2010;37:622132. 25. McCarville MB, Lederman HM, Santana VM, Daw NC, Shochat SJ, Li CS, et al. Distinguishing benign from malignant pulmonary nodules with helical chest CT in children with malignant solid tumors. Radiology. 2006;239:51420. 26. Cistaro A, Lopci E, Gastaldo L, Fania P, Brach Del Prever A, Fagioli F. The role of 18F-FDG PET/CT in the metabolic characterization of lung nodules in pediatric patients with bone sarcoma. Pediatr Blood Cancer. 2012;59:120610.

11.

12.

13.

14.

15.

16.

17.