Abstract
Background
Skeletal metastases of bone sarcomas are indicators of poor prognosis. Various imaging modalities are available for their identification, which include bone scan, positron emission tomography/CT scan, MRI, and bone marrow aspiration/biopsy. However, there is considerable ambiguity regarding the best imaging modality to detect skeletal metastases. To date, we are not sure which of these investigations is best for screening of skeletal metastasis.
Question/purpose
Which staging investigation—18F-fluorodeoxyglucose positron emission tomography/CT (18F-FDG PET/CT), whole-body MRI, or 99mTc-MDP skeletal scintigraphy—is best in terms of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) in detecting skeletal metastases in patients with osteosarcoma and those with Ewing sarcoma?
Methods
A prospective diagnostic study was performed among 54 of a total 66 consecutive osteosarcoma and Ewing sarcoma patients who presented between March 2018 and June 2019. The institutional review board approved the use of all three imaging modalities on each patient recruited for the study. Informed consent was obtained after thoroughly explaining the study to the patient or the patient’s parent/guardian. The patients were aged between 4 and 37 years, and their diagnoses were proven by histopathology. All patients underwent 99mTc-MDP skeletal scintigraphy, 18F-FDG PET/CT, and whole-body MRI for the initial staging of skeletal metastases. The number and location of bone and bone marrow lesions diagnosed with each imaging modality were determined and compared with each other. Multidisciplinary team meetings were held to reach a consensus about the total number of metastases present in each patient, and this was considered the gold standard. The sensitivity, specificity, PPV, and NPV of each imaging modality, along with their 95% confidence intervals, were generated by the software Stata SE v 15.1. Six of 24 patients in the osteosarcoma group had skeletal metastases, as did 8 of 30 patients in the Ewing sarcoma group. The median (range) follow-up for the study was 17 months (12 to 27 months). Although seven patients died before completing the minimum follow-up, no patients who survived were lost to follow-up.
Results
With the number of patients available, we found no differences in terms of sensitivity, specificity, PPV, and NPV among the three staging investigations in patients with osteosarcoma and in patients with Ewing sarcoma. Sensitivities to detect bone metastases for 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy were 100% (6 of 6 [95% CI 54% to 100%]), 83% (5 of 6 [95% CI 36% to 100%]), and 67% (4 of 6 [95% CI 22% to 96%]) and specificities were 100% (18 of 18 [95% CI 82% to 100%]), 94% (17 of 18 [95% CI 73% to 100%]), and 78% (14 of 18 [95% CI 52% to 94%]), respectively, in patients with osteosarcoma. In patients with Ewing sarcoma, sensitivities to detect bone metastases for 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy were 88% (7 of 8 [95% CI 47% to 100%]), 88% (7 of 8 [95% CI 47% to 100%]), and 50% (4 of 8 [95% CI 16% to 84%]) and specificities were 100% (22 of 22 [95% CI 85% to 100%]), 95% (21 of 22 [95% CI 77% to 100%]), and 95% (21 of 22 [95% CI 77% to 100%]), respectively. Further, the PPVs for detecting bone metastases for 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy were 100% (6 of 6 [95% CI 54% to 100%]), 83% (5 of 6 [95% CI 36% to 100%]), and 50% (4 of 8 [95% CI 16% to 84%]) and the NPVs were 100% (18 of 18 [95% CI 82% to 100%]), 94% (17 of 18 [95% CI 73% to 100%]), and 88% (14 of 16 [95% CI 62% to 98%]), respectively, in patients with osteosarcoma. Similarly, the PPVs for detecting bone metastases for 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy were 100% (7 of 7 [95% CI 59% to 100%]), 88% (7 of 8 [95% CI 50% to 98%]), and 80% (4 of 5 [95% CI 28% to 100%]), and the NPVs were 96% (22 of 23 [95% CI 78% to 100%]), 95% (21 of 22 [95% CI 77% to 99%]), and 84% (21 of 25 [95% CI 64% to 96%]), respectively, in patients with Ewing sarcoma. The confidence intervals around these values overlapped with each other, thus indicating no difference between them.
Conclusion
Based on these results, we could not demonstrate a difference in the sensitivity, specificity, PPV, and NPV between 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy for detecting skeletal metastases in patients with osteosarcoma and Ewing sarcoma. For proper prognostication, a thorough metastatic workup is essential, which should include a highly sensitive investigation tool to detect skeletal metastases. However, our study findings suggest that there is no difference between these three imaging tools. Since this is a small group of patients in whom it is difficult to make broad recommendations, these findings may be confirmed by larger studies in the future.
Level of Evidence
Level II, diagnostic study.
Introduction
Primary bone malignancies account for fewer than 0.2% of all cancers; osteosarcoma and Ewing sarcoma are the two most common neoplasms [4]. According to the National Cancer Institute’s Surveillance, Epidemiology, and End Results program database, the incidence of these primary bone malignancies has increased in the past four decades from 0.82 patients per 1 million in 1975 to 0.91 per 1 million in 2011 [13]. However, the 5-year survival rate of these tumors has dramatically increased during the past four decades, from 10% to 20% in the 1970s to 70% in 2011 [1, 13]. Although the data show an improvement in survival, and despite treatment advancements, the prognosis for these patients is still poor. The most important survival predictor is the presence or absence of metastasis at presentation [11, 12]. The lung is the most common metastasis site in patients with osteosarcoma and Ewing sarcoma, followed by bone metastases, which occur with a frequency of 14% to 83% [7, 11, 12]. This hematogenous spread can occur early and adversely affect prognosis. Identifying disease sites and locating metastases affects staging and the initial risk stratification, and it is particularly important for Ewing sarcoma because all associated disease sites can receive local therapy [6]. Therefore, the initial workup is important for proper management.
Staging investigations after an osteosarcoma diagnosis, according to current National Comprehensive Cancer Network guidelines [14], include chest radiographs, 99mTechnetium-methylene diphosphonate (99mTc-MDP) skeletal scintigraphy (bone scan), noncontrast CT (NCCT) of the chest, and local MRI. In patients with Ewing sarcoma, in addition to these, bone marrow biopsy is also included in the staging workup to identify marrow involvement [14]. 99mTc-MDP skeletal scintigraphy is the most commonly used method for detecting skeletal metastases [9]. Historically, the ability of skeletal scintigraphy to identify skeletal metastases when they are clinically and radiographically occult is unequalled by other imaging techniques [2]. However, osteolytic and marrow lesions can easily be overlooked, as in early stages of Ewing sarcoma. 18F-fluorodeoxyglucose positron emission tomography/CT (18F-FDG PET/CT) is useful in the staging workup, especially in Ewing sarcoma because it can detect skeletal metastases before an osteoblastic host response develops. Recently, whole-body MRI has been a useful tool in staging, especially in small-cell neoplasms such as Ewing sarcoma [8]. Diffusion-weighted MRI provides functional information and can be used to detect and characterize pathologic processes, which may therefore be valuable in staging and follow-up imaging of malignant tumors [10]. However, the role of this modality in staging workup is yet to be defined. For lung metastases, NCCT of the chest is the standard modality. However, for bone metastases, the modality of choice is still debatable. Each of the available staging methods provides complementary prognostic information; however, the optimal combination of these staging methods is unclear [9].
We therefore asked: Which staging investigation—skeletal scintigraphy, whole-body MRI, or 18F-FDG PET/CT—is best in terms of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) in detecting skeletal metastases in patients with osteosarcoma and those with Ewing sarcoma?
Patients and Methods
Study Design
This prospective diagnostic study was performed at our institution from March 2018 to June 2019, and the patients were followed for a minimum of 12 months to a maximum of 27 months (median 17 months).
Participants
The study included 54 patients; 44% (24) had osteosarcoma and 56% (30) had Ewing sarcoma. From March 2018 to June 2019, patients with histopathologically proven osteosarcoma and those with Ewing sarcoma were recruited for this study, regardless of the age or sex of the patient or the location of the tumor (axial or appendicular). A total of 66 patients presented to our institute during the study period, and 12 were excluded (6% [4 of 66] refused to participate and 12% [8 of 66] missed one or the other investigations) (Fig. 1).
Fig. 1.
A Standards for Reporting of Diagnostic Accuracy Studies (STARD) flow diagram that demonstrates patient recruitment during the study period.
Description of Study Population
The study population was divided into two groups: patients with osteosarcoma and those with Ewing sarcoma. The mean age of patients with osteosarcoma was 17 ± 7 years. The mean age of patients with Ewing sarcoma was 15 ± 8 years (Table 1). Seventy percent (38 of 54) of the recruited patients were males and 30% (16 of 54) were females. Among the patients with osteosarcoma, 75% (18 of 24) were males and 25% (6 of 24) were females, while among the patients with Ewing sarcoma, 67% (20 of 30) were males and 33% (10 of 30) were females (self-reported gender and not biologic sex) (Table 1). With the number of patients we had, we could not detect a difference in the age and sex distribution of patients between the two groups (p > 0.05). Twelve percent (8 of 66) of patients missed their appointment for one scan or the other (incomplete study) and had to be excluded as rescheduling of appointment for bone scan, PET, or MRI at our institute is quite difficult due to long waitlists for the scans. Except for the seven patients who died before completing the minimum follow-up, no other patients were lost to follow-up before 12 months. The patients excluded from the study were not different in terms of age, sex, or diagnosis from the group of patients recruited for the study.
Table 1.
Age and sex distribution in the study population
| Parameter | Osteosarcoma (n = 24) | Ewing sarcoma (n = 30) | p value |
| Age in years, mean ± SD | 17 ± 7 | 15 ± 8 | 0.34 |
| Sex, % (n) Male Female |
75 (18) 25 (6) |
67 (20) 33 (10) |
0.56 |
Ours is a public sector teaching hospital, and we have a system of exempting charges of investigations done for thesis purposes. Hence, no patients had to pay for any investigations, and no patients dropped out due to the inability to bear the costs of the additional studies.
Test Methods
All patients underwent all three modalities of investigation, that is, 99mTc-MDP skeletal scintigraphy, whole-body MRI, and 18F-FDG PET/CT within 1 week to detect skeletal metastases (with at least 24 hours between 99mTc-MDP skeletal scintigraphy and 18F-FDG PET/CT), in a random order as an initial staging procedure after diagnosis of the tumor.
FDG PET/CT Scanning
18F-FDG PET/CT was performed on a PET/CT scanner (64-slice, Siemens®; and 128-slice GE® PET/CT scanner). Patients fasted for at least 4 hours before the test to ensure that blood glucose levels were < 150 mg/dL. Immediately before 18F-FDG PET/CT, patients were asked to empty their bladder. Scanning began 1 hour after the intravenous administration of 0.142 mCi/kg of 18F-FDG. Low-dose CT was followed by whole-body PET acquisition. PET images were examined for areas of increased radiotracer uptake. The corresponding areas on CT and fused PET/CT images were corroborated. Any focal area of uptake more than that of the uptake by the liver was considered metastatic.
Whole-body MRI
MRI was performed on a 1.5-T Phillips Achieva® whole-body MR scanner (Philips Healthcare). The patients underwent MRI in the supine position with the arms placed next to the thorax and abdomen using a single-coil (Q coil) system. The smallest possible coils were used for very young patients. Diffusion-weighted images with background body signal suppression and short tau inversion recovery sequences were taken. In the diffusion-weighted imaging with background body signal suppression sequence, 6-mm slices with 1-mm overlap were taken in axial sections. The repetition time was 7000 and the echo time was 80. The time of inversion was 150 and the b value was 800 s/mm2. In the short tau inversion recovery sequence, 5-mm slices were taken in coronal sections. The repetition time was 6000, the echo time was 60, and the time of inversion was 150. In both sequences, eight or nine stations were scanned, and the number of slices depended on the patient’s size. On diffusion-weighted images with background body signal suppression and short tau inversion recovery images, a metastatic bone or bone marrow lesion was seen as focal or diffuse hyperintense bone marrow signal intensity relative to the adjacent (or, in the extremities, contralateral) normal bone marrow.
99mTc-MDP Skeletal Scintigraphy
99mTc-MDP skeletal scintigraphy was performed 4 hours after the injection of 99mTc-MDP using a double-head Gamma camera (Discovery NM 670, GE Healthcare). Delayed static images were acquired 4 hours after the radiotracer injection. Multiple overlapping spot images were obtained over the entire body, including the head, trunk, and extremities, in an average time of 10 minutes. The radiotracer dose was adjusted according to the patient’s body weight. A metastatic bone lesion was defined as an area of focally increased radionuclide uptake relative to adjacent and contralateral normal tissue that was not located in a region of physiologically increased uptake.
Interpretation of Imaging Findings
MR images were evaluated by an experienced radiologist (SG), and the skeletal scintigrams and 18F-FDG PET/CT images were analyzed by an experienced nuclear medicine physician (SAS). The two experts were unaware of the results of the other modality. Additionally, the nuclear medicine physician evaluated the skeletal scintigrams and 18F-FDG PET/CT images of the patients in a blinded manner. Blinding was done by getting the patients’ 18F-FDG PET/CT and skeletal scintigraphy images reported to the nuclear medicine physician at different timepoints in a random order. The number and location of metastatic lesions were determined for all three imaging modalities in each patient (Fig. 2). After imaging was completed, we revealed the individual findings of the three imaging techniques to the examiners in a multidisciplinary team meeting, and the final reports were confirmed. No radiographs were done for individual metastatic sites in any patient. Ideally, a histopathologic examination of all suspected skeletal metastatic sites should be considered as the gold standard investigation for confirming the metastases. However, due to the multiplicity and the deep location of metastatic sites, such exhaustive and invasive investigations cannot be ethically justified. Also, biopsy confirmation of only some selected lesions in patients with multiple metastatic lesions might be considered suboptimal. Instead, multidisciplinary team consensus by experts was considered as the gold standard for the purpose of this study. Whenever a consensus was not achievable on a particular lesion, an interval PET/CT scan was performed after completion of preoperative chemotherapy cycles (interval PET/CT scan was done for two patients with Ewing sarcoma and none with osteosarcoma). In lesions that previously showed increased uptake and were therefore suspected of being metastatic, a decrease in FDG uptake in the lesions on follow-up 18F-FDG PET/CT after chemotherapy confirmed that the lesions were actually metastatic (Fig. 3). The false-positive lesions were mostly incidental cartilage lesions, which were confirmed to be so on conventional MRI (T1- and T2-weighted images) of the area (Fig. 4). Thus, a final call on the lesion as whether true positive or false positive was taken based on a repeat multidisciplinary team discussion (Fig. 5).
Fig. 2.
A-D These (A) anterior-view and (B) posterior-view 99mTc-MDP whole-body bone scan images are from an 11-year-old girl with Ewing sarcoma of the right distal femur and show heterogeneous radiotracer uptake in a mass in the right distal femur with no other metastatic lesion (false-negative bone scan). (C) A maximum-intensity projection of an 18F-FDG PET/CT scan shows heterogeneous uptake in a large expansile mass in the right distal thigh along with increased uptake suggesting a metastasis in the left scapula; left proximal humerus; D11, D12, and L4 vertebrae; left ilium; right ischium; right acetabulum; and right proximal tibia. (D) An inverted (negative) image of a coronal diffusion-weighted whole-body MR image of the same patient shows a hypointense lesion in the right distal femur and surrounding soft tissues of the distal thigh, suggestive of a primary tumor, along with metastases in the left scapula, left proximal humerus, D12 and L4 vertebrae, left ilium, right ischium, right acetabulum, and right proximal tibia. It only failed to detect one metastatic lesion of the D11 vertebra compared with 18F-FDG PET/CT.
Fig. 3.
A-H These transaxial fused 18F-FDG PET/CT images from a 10-year-old boy with Ewing sarcoma of the right fibula demonstrate areas of increased FDG uptake in the (A) shaft of the right femur, (B) left sacral ala, (C) left head of the femur, and (D) left iliac bone, suggesting metastatic lesions. Transaxial fused 18F-FDG PET/CT images of this patient at the same four levels 5 months after chemotherapy demonstrate decreased FDG uptake in the (E) shaft of the right femur, (F) left sacral ala, (G) left head of the femur, and (H) left iliac bone, which were previously showing high FDG uptake, confirming that these lesions were actually skeletal metastases from the primary tumor. This also gave us information that the tumor was responding to the given chemotherapeutic regimen, highlighting the utility of 18F-FDG PET/CT in assessing the patient’s response to treatment.
Fig. 4.
A-D (A) A 99mTc-MDP whole-body bone scan of a 10-year-old boy with an osteosarcoma of the left distal femur shows increased uptake in the ipsilateral proximal tibia, suggesting a metastatic lesion (arrow). (B) A maximum-intensity projection of an 18F-FDG PET scan shows heterogeneous radiotracer uptake in the distal part of the left thigh (primary tumor), with no other areas of abnormal radiotracer uptake. (C) A coronal short tau inversion recovery whole-body MR image of the same patient shows a lesion in the left proximal tibia. However, this lesion had an altered signal intensity, dissimilar to the primary lesion in the distal femur. (D) An additional T2-weighted spin-echo-sequence MRI of the left knee was performed in which the proximal tibia lesion was confirmed to be an enchondroma, rather than an osseous metastasis from the primary tumor.
Fig. 5.
This flow chart shows the study protocol.
Ethical Approval
Ethical approval for this study was obtained from the Institute Ethics Committee for Post Graduate Research All India Institute of Medical Sciences (Ref. no.: IECPG-626/31.01.2018, RT-21/28.02.2018).
Statistical Analysis
The number of patients with or without skeletal metastases as detected by each imaging modality was summarized in frequency tables. For each patient with osteosarcoma, all three imaging modalities were characterized as true-positives, false-positives, false-negatives, and true-negatives (Table 2). The same was done for the Ewing sarcoma patients (Table 3). For example, if the multidisciplinary team consensus concluded that the patient had at least one skeletal metastasis and PET/CT scan and MRI also detected at least one skeletal metastasis, but the skeletal scintigraphy did not detect any, then PET/CT scan and MRI were true positive and skeletal scintigraphy was false negative for this patient. Similarly, if the multidisciplinary team meeting concluded that the patient did not have any skeletal metastases and PET/CT and whole-body MRI were also negative for skeletal metastases, but skeletal scintigraphy showed a suspected metastasis at a particular site and that lesion was later proven to be a benign enchondroma in a T1/T2-weighted local MRI of that area, then in that case, PET/CT and whole-body MRI were considered true negative for that patient and skeletal scintigraphy was false positive for that patient. It is important here to note that individual lesions were not classified as true or false positives or negatives but this was done for a patient as a whole. In both tumor groups, for each imaging modality, the sensitivity, specificity, PPV, and NPV were calculated along with their 95% confidence intervals. All statistical calculations were done in STATA/SE version 15.1.
Table 2.
Comparison of all three imaging modalities in the osteosarcoma group (n = 24)
| Imaging modality | Report of the imaging modality | Number of patients with metastasis | p values | ||||
| Present | Absent | Total | vs bone scan | vs PET/CT scan | vs whole-body MRI | ||
| Bone scan | Positive | 4 | 4 | 8 | |||
| Negative | 2 | 14 | 16 | ||||
| Total | 6 | 18 | 24 | ||||
| Sensitivity (95% CI) | 67% (22%-96%) | 0.44 | 0.51 | ||||
| Specificity (95% CI) | 78% (52%-94%) | 0.11 | 0.34 | ||||
| PPV (95% CI) | 50% (16%-84%) | ||||||
| NPV (95% CI) | 88 (62%-98%) | ||||||
| Present | Absent | Total | |||||
| PET/CT scan | Positive | 6 | 0 | 6 | |||
| Negative | 0 | 18 | 18 | ||||
| Total | 6 | 18 | 24 | ||||
| Sensitivity (95% CI) | 100% (54%-100%) | 0.44 | 0.30 | ||||
| Specificity (95% CI) | 100% (82%-100%) | 0.11 | 0.31 | ||||
| PPV (95% CI) | 100% (54%-100%) | ||||||
| NPV (95% CI) | 100% (82%-100%) | ||||||
| Present | Absent | Total | |||||
| Whole-body MRI | Positive | 5 | 1 | 6 | |||
| Negative | 1 | 17 | 18 | ||||
| Total | 6 | 18 | 24 | ||||
| Sensitivity (95% CI) | 83% (36%-100%) | 0.51 | 0.30 | ||||
| Specificity (95% CI) | 94% (73%-100%) | 0.34 | 0.31 | ||||
| PPV (95% CI) | 83% (36%-100%) | ||||||
| NPV (95% CI) | 94% (73%-100%) | ||||||
PPV = positive predictive value; NPV = negative predictive value.
Table 3.
Comparison of all three imaging modalities in the Ewing sarcoma group (n = 30)
| Imaging modality | Report of the imaging modality | Number of patients with metastasis | p values | ||||
| Present | Absent | Total | vs bone scan | vs PET/CT scan | vs whole-body MRI | ||
| Bone scan | Positive | 4 | 1 | 5 | |||
| Negative | 4 | 21 | 25 | ||||
| Total | 8 | 22 | 30 | ||||
| Sensitivity (95% CI) | 50% (16%-84%) | 0.28 | 0.28 | ||||
| Specificity (95% CI) | 95% (77%-100%) | 0.31 | > 0.99 | ||||
| PPV (95% CI) | 80% (28%-100%) | ||||||
| NPV (95% CI) | 84% (64%-96%) | ||||||
| Present | Absent | Total | |||||
| PET/CT scan | Positive | 7 | 0 | 7 | |||
| Negative | 1 | 22 | 23 | ||||
| Total | 8 | 22 | 30 | ||||
| Sensitivity (95% CI) | 88% (47%-100%) | 0.28 | > 0.99 | ||||
| Specificity (95% CI) | 100% (85%-100%) | 0.31 | 0.31 | ||||
| PPV (95% CI) | 100% (59%-100%) | ||||||
| NPV (95% CI) | 96% (78%-100%) | ||||||
| Present | Absent | Total | |||||
| Whole-body MRI | Positive | 7 | 1 | 8 | |||
| Negative | 1 | 21 | 22 | ||||
| Total | 8 | 22 | 30 | ||||
| Sensitivity (95% CI) | 88% (47%-100%) | 0.28 | > 0.99 | ||||
| Specificity (95% CI) | 95% (77%-100%) | > 0.99 | 0.31 | ||||
| PPV (95% CI) | 88% (50%-98%) | ||||||
| NPV (95% CI) | 95% (77%-99%) | ||||||
PPV = positive predictive value; NPV = negative predictive value.
Results
Which Staging Investigation Is Best in Detecting Skeletal Metastases from Osteosarcoma And Ewing Sarcoma?
With the number of patients available, we found no differences in terms of sensitivity, specificity, PPV, and NPV among the three staging investigations in both osteosarcoma (Table 2) and Ewing sarcoma patients (Table 3).
Sensitivity
In patients with osteosarcoma, the sensitivity for detecting skeletal metastases with 18F-FDG PET/CT was 100% (6 of 6 [95% CI 54% to 100%]), with whole-body MRI it was 83% (5 of 6 [95% CI 36% to 100%]), and with 99mTc-MDP skeletal scintigraphy it was 67% (4 of 6 [95% CI 22% to 96%]) (PET/CT versus MRI p = 0.30, PET/CT versus skeletal scintigraphy p = 0.44, and MRI versus skeletal scintigraphy p = 0.51).
Similarly, in patients with Ewing sarcoma, the sensitivity of 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy for detecting skeletal metastases was 88% (7 of 8 [95% CI 47% to 100%]), 88% (7 of 8 [95% CI 47% to 100%]), and 50% (4 of 8 [95% CI 16% to 84%]), respectively (PET/CT versus MRI p > 0.99, PET/CT versus skeletal scintigraphy p = 0.28, and MRI versus skeletal scintigraphy p = 0.28).
Specificity
For detecting skeletal metastases in patients with osteosarcoma, the specificity of 18F-FDG PET/CT was 100% (18 of 18 [95% CI 82% to 100%]), for whole-body MRI it was 94% (17 of 18 [95% CI 73% to 100%]), and for 99mTc-MDP skeletal scintigraphy it was 78% (14 of 18 [95% CI 52% to 94%]) (PET/CT versus MRI p = 0.31, PET/CT versus skeletal scintigraphy p = 0.11, and MRI versus skeletal scintigraphy p = 0.34).
On the other hand, the specificities of 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy for detecting skeletal metastases in patients with Ewing sarcoma were 100% (22 of 22 [95% CI 85% to 100%]), 95% (21 of 22 [95% CI 77% to 100%]), and 95% (21 of 22 [95% CI 77% to 100%]), respectively (PET/CT versus MRI p = 0.31, PET/CT versus skeletal scintigraphy p = 0.31, and MRI versus skeletal scintigraphy p > 0.99).
Positive Predictive Value
The PPVs of 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy for detecting skeletal metastases in patients with osteosarcoma were 100% (6 of 6 [95% CI 54% to 100%]), 83% (5 of 6 [95% CI 36% to 100%]), and 50% (4 of 8 [95% CI 16% to 84%]), respectively.
For detecting skeletal metastases in patients with Ewing sarcoma, the PPV of 18F-FDG PET/CT was 100% (7 of 7 [95% CI 59% to 100%]), it was 88% (7 of 8 [95% CI 50% to 98%]) for whole-body MRI, and it was 80% (4 of 5 [95% CI 28% to 100%]) for 99mTc-MDP skeletal scintigraphy.
Negative Predictive Value
For detecting skeletal metastases in patients with osteosarcoma, the NPVs were 100% (18 of 18 [95% CI 82% to 100%]) for 18F-FDG PET/CT, 94% (17 of 18 [95% CI 73% to 100%]) for whole-body for MRI, and 88% (14 of 16 [95% CI 62% to 98%]) for 99mTc-MDP skeletal scintigraphy.
The NPVs of 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy for detecting skeletal metastases in patients with Ewing sarcoma were 96% (22 of 23 [95% CI 78% to 100%]), 95% (21 of 22 [95% CI 77% to 99%]), and 84% (21 of 25 [95% CI 64% to 96%]), respectively.
The confidence intervals around these values overlapped with each other, thus indicating no difference between them.
Discussion
Conventionally, the standard evaluation for distant metastases of bone sarcomas has included 99mTc-MDP skeletal scintigraphy, NCCT of the chest, and in addition, bone marrow aspiration for Ewing sarcoma. However, whole-body imaging, such as 18F-FDG PET or whole-body MRI, has been recently added for identifying distant disease sites. Since NCCT is better than 18F-FDG PET for detecting lung metastases smaller than 8 mm [15], chest CT remains an essential part of the staging process. However, there is no universal consensus on the best imaging modality to detect skeletal metastases. Bone scintigraphy has been thought to be better for detecting skeletal metastases in patients with osteosarcomas, whereas involvement of the cortex with osteoblastic activity occurs early in patients in whom skeletal metastases develop. In contrast, in Ewing sarcoma, which is primarily a pathologic condition of the marrow, the osteoblastic response occurs later than in osteosarcoma. Thus, 18F-FDG PET/CT and MRI have been thought to be better tools because they can detect metastases of the skeletal marrow before an osteoblastic host response develops. In this study, we could not find any difference in the diagnostic ability of 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy in detecting skeletal metastases in patients with osteosarcoma and Ewing sarcoma.
Limitations
Ideally, a histopathologic confirmation of all suspected metastatic sites should be done to confirm whether they are actually metastases. A limitation of the study is that histopathologic confirmation was not available for metastatic lesions. However, due to multiplicity and deep location of metastatic sites, such an invasive and exhaustive modality could not be ethically justified. Previously, Kumar et al. [8] have used similar multidisciplinary discussions as the gold standard for deciding on skeletal metastasis because performing a biopsy of every lesion would cause substantial morbidity to the patient. Also, follow-up imaging was not done for every patient in this study and was only reserved for situations with discrepancy in multidisciplinary team discussions. Performing follow-up imaging in all patients may have increased our diagnostic accuracy; however, it would have increased the cost for the institute and would have led to additional radiation exposure to all patients. We did not do an interobserver correlation analysis. However, our radiology and nuclear medicine experts have had more than a decade of expertise specializing in musculoskeletal oncology. Another limitation is the availability of these imaging modalities, and the fact that these advanced imaging modalities require considerable experience in reporting, which may not be available in all centers worldwide. Since our study population mostly included adolescents and young adults, our findings may be limited to this population only. Nevertheless, the two sarcomas studied here most commonly occur in this age group. Also, the sex ratio of our study sample is not proportionate. Therefore, our study results may not be generalizable to both sexes, especially to the female population. Finally, our patient numbers are insufficient because this research was performed as part of a thesis study, and thus, some of the comparisons might have been statistically valid if the study was done in larger numbers of patients.
Which Staging Investigation Is Best in Detecting Skeletal Metastases from Osteosarcoma and Ewing Sarcoma?
Although the absolute values of the sensitivity, specificity, PPV, and NPV in both groups were different, the CIs around these values were wide and overlapped with each other for all three diagnostic investigations. Therefore, with the limited number of patients included in our study, we could not demonstrate a difference between the sensitivity, specificity, PPV, and NPV of 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy for detecting skeletal metastases in patients with osteosarcoma and those with Ewing sarcoma. This indicates that any of the three investigations can be used for the initial staging of patients with osteosarcoma and those with Ewing sarcoma to categorize the patient as with or without skeletal metastases.
To the best of our knowledge, only two studies have compared all three imaging modalities for detecting skeletal metastases. In a study by Daldrup-Link et al. [3], among 39 children with bone and soft tissue sarcomas, the sensitivities for detecting bone metastases were 90% for 18F-FDG PET/CT, 82% for whole-body MRI, and 71% for 99mTc-MDP skeletal scintigraphy. However, only three patients in the cohort had osteosarcoma, and the authors included multiple other bone and soft tissue tumors such as rhabdomyosarcoma, lymphoma, myeloma, malignant melanoma, and Langerhans cell histiocytosis. Kumar et al. [8] reported that whole-body MRI revealed metastases in 39 of 208 regions in 26 patients, 99mTc-MDP skeletal scintigraphy in 12 regions, and 18F-FDG PET/CT in 36 regions. The authors concluded that whole-body MRI and 18F-FDG PET/CT were excellent imaging modalities for screening for skeletal metastases and were far more accurate than 99mTc-MDP skeletal scintigraphy. However, this study was performed in a cohort of patients with small-cell neoplasms of bone and soft tissues such as neuroblastoma, primitive neuroectodermal tumor, rhabdomyosarcoma, and Ewing sarcoma; it did not include patients with osteosarcoma.
Also, a prospective multicenter study by Franzius et al. [5] showed that 18F-FDG PET performed better than planar bone scan in detecting osseous metastases in Ewing sarcoma. The authors compared 18F-FDG PET with 99mTc-MDP skeletal scintigraphy and other conventional imaging modalities, and they found the sensitivity of 18F-FDG PET to be 90% and that of 99mTc-MDP skeletal scintigraphy to be 81% in identifying osseous metastases [15]. The authors demonstrated 100% sensitivity when images from all modalities were compared in a side-by-side analysis. 18F-FDG PET was 91% accurate for locating distant metastases compared with 47% accuracy with conventional imaging. Furthermore, the authors confirmed that the findings obtained by performing an 18F-FDG PET scan led to changes in management in terms of local control or chemotherapy in approximately half of the patients. However, in our study, we found no difference among 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy in terms of sensitivity, specificity, PPV, and NPV in detecting metastatic bone disease. Also, there are some downsides to using PET/CT scan due to the high cost and the increased radiation exposure and also to using whole-body MRI again because of the cost and time taken to perform the scan, compared with the current standard investigation for skeletal metastases (skeletal scintigraphy).
Conclusion
In our study, with the numbers of patients we had, we could not demonstrate a difference between the sensitivity, specificity, PPV, and NPV of 18F-FDG PET/CT, whole-body MRI, and 99mTc-MDP skeletal scintigraphy for detecting skeletal metastases in patients with osteosarcoma and those with Ewing sarcoma. Nevertheless, for proper prognostication, a thorough metastatic workup is essential and should include a highly sensitive investigation tool to detect skeletal metastases. Since this is a small group of patients in whom it is difficult to make broad recommendations, future research is needed to better define the complementary use of 18F-FDG PET/CT, 99mTc-MDP skeletal scintigraphy, and whole-body MRI in patients with musculoskeletal malignancies.
Acknowledgments
We thank Roshan Banjara MS, of the Department of Orthopaedics at the All India Institute of Medical Sciences, New Delhi, India, for helping with data collection.
Footnotes
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.
Each author certifies that neither he, nor any member of his immediate family, has funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article.
Clinical Orthopaedics and Related Research® neither advocates nor endorses the use of any treatment, drug, or device. Readers are encouraged to always seek additional information, including FDA approval status, of any drug or device before clinical use.
Ethical approval for this study was obtained from the Institute Ethics Committee for Post Graduate Research All India Institute of Medical Sciences (Ref. no.: IECPG-626/31.01.2018, RT-21/28.02.2018).
References
- 1.Anninga JK, Gelderblom H, Fiocco M, et al. Chemotherapeutic adjuvant treatment for osteosarcoma: where do we stand? Eur J Cancer. 2011;47:2431-2445. [DOI] [PubMed] [Google Scholar]
- 2.Connolly LP, Drubach LA, Treves TS. Applications of nuclear medicine in pediatric oncology. Clin Nucl Med. 2002;27:117-125. [DOI] [PubMed] [Google Scholar]
- 3.Daldrup-Link HE, Franzius C, Link TM, et al. Whole-body MR imaging for detection of bone metastases in children and young adults: comparison with skeletal scintigraphy and FDG PET. Am J Roentgenol. 2001;177:229-236. [DOI] [PubMed] [Google Scholar]
- 4.Dean BJ, Whitwell D. Epidemiology of bone and soft-tissue sarcomas. Orthop Trauma. 2009;23:223-230. [Google Scholar]
- 5.Franzius C, Sciuk J, Daldrup-Link HE, Jürgens H, Schober O. FDG-PET for detection of osseous metastases from malignant primary bone tumours: comparison with bone scintigraphy. Eur J Nucl Med. 2000;27:1305-1311. [DOI] [PubMed] [Google Scholar]
- 6.Furth C, Amthauer H, Denecke T, Ruf J, Henze G, Gutberlet M. Impact of whole-body MRI and FDG-PET on staging and assessment of therapy response in a patient with Ewing sarcoma. Pediatr Blood Cancer. 2006;47:607-611. [DOI] [PubMed] [Google Scholar]
- 7.Jeffree GM, Price CH, Sissons HA. The metastatic patterns of osteosarcoma. Br J Cancer. 1975;32:87-107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kumar J, Seith A, Kumar A, et al. Whole-body MR imaging with the use of parallel imaging for detection of skeletal metastases in pediatric patients with small-cell neoplasms: comparison with skeletal scintigraphy and FDG PET/CT. Pediatr Radiol. 2008;38:953-962. [DOI] [PubMed] [Google Scholar]
- 9.Kundu ZS. Classification, imaging, biopsy and staging of osteosarcoma. Indian J Orthop. 2014;48:238-246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lauenstein TC, Goehde SC, Herborn CU, et al. Whole-body MR imaging: evaluation of patients for metastases. Radiology. 2004;233:139-148. [DOI] [PubMed] [Google Scholar]
- 11.López Guerra JL, Márquez-Vega C, Ramírez-Villar GL, et al. Prognostic factors for overall survival in paediatric patients with Ewing sarcoma of bone treated according to multidisciplinary protocol. Clin Transl Oncol. 2012;14:294-301. [DOI] [PubMed] [Google Scholar]
- 12.Mialou V, Philip T, Kalifa C, et al. Metastatic osteosarcoma at diagnosis: prognostic factors and long-term outcome—the French pediatric experience. Cancer. 2005;104:1100-1109. [DOI] [PubMed] [Google Scholar]
- 13.National Cancer Institute. SEER stat fact sheet. Cancer: soft tissue. Available at: http://seer.cancer.gov/statfacts/html/bones.html. Accessed November 21, 2019.
- 14.National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: bone cancer. Available at: http://www.nccn.org/professionals/physician_gls/pdf/bone_blocks.pdf. Accessed November 21, 2019.
- 15.Völker T, Denecke T, Steffen I, et al. Positron emission tomography for staging of pediatric sarcoma patients: results of a prospective multicenter trial. J Clin Oncol. 2007;25:5435-5441. [DOI] [PubMed] [Google Scholar]