Abstract
Background/Aim: Surgical staging is paramount to treatment of primary bone sarcomas. Often, bone scintigraphy and/or positron emission tomography-computed tomography (PET-CT) are used to exclude skeletal metastases; however, skeletal metastases in chondrosarcoma are rare. The purpose of this study was to assess the utility of these staging methods in patients with chondrosarcoma. Patients and Methods: We reviewed 138 (87 males, 51 female) patients, mean age 54±20 years, with a chondrosarcoma, who had completed a bone scintigraphy or PET/CT as part of surgical staging. Sensitivity, specificity, and positive/negative predictive value of the scans was calculated. Results: Seventeen (12%) patients had a positive bone scintigraphy or PET-CT for skeletal metastases. In cases of bone scintigraphy (n=11), 6 were benign and 5 were skeletal metastases. In cases of PET-CT, 6 were skeletal metastases, 3 were positive and 3 benign. All positive cases regarded dedifferentiated chondrosarcoma. The overall sensitivity and specificity of a bone scan or PET-CT was 100% and 93.1%; with a positive and negative predictive value of 47.1% and 100%, respectively. Conclusion: Skeletal metastases at presentation of chondrosarcoma are rare and associated with dedifferentiated chondrosarcoma. Bone scintigraphy or PET-CT should only be performed in cases of high grade and dedifferentiated histology.
Chondrosarcoma is the second most common primary malignant bone tumor, often affecting adults (1, 2). As in all primary bone sarcomas, proper surgical staging is paramount in these cases given implications for both treatment and prognosis. Traditionally, investigations have included magnetic resonance imaging (MRI) to define local tumor extent, chest computed tomography (CT) to exclude pulmonary metastases, and bone scintigraphy or positron emission tomography (PET) CT to exclude skeletal metastases, a practice largely based on osteosarcomas (3-5). However, review of available literature on series of patients with chondrosarcoma of the bone suggests that the presence of skeletal metastases at diagnosis is exceedingly rare (1, 5-7).
Although sensitive, bone scintigraphy and PET-CT often have a low specificity, resulting in the need for additional imaging and/or investigation to exclude benign entities (8, 9). This leads one to question the utility of routine bone scintigraphy and PET-CT in the initial staging these of patients. Therefore, the purpose of this study was to assess the utility of bone scintigraphy or PET-CT in the surgical staging of patients surgically treated for chondrosarcoma of the bone at our institution.
Patients and Methods
Following Institutional Review Board (IRB) approval, we retrospectively reviewed our institutional sarcoma database to identify patients with histologically confirmed chondrosarcoma at our institution from 1998-2017. Those with skull base, head and neck, extraskeletal or recurrent disease were excluded. Ultimately, 276 patients with chondrosarcoma of the bone were included in the study. Of these, 138 (50%) patients underwent advanced imaging for evaluation of bone metastatic disease. This included 119 (86%) patients who had a bone scan and 29 (24%) patients who had a PET-CT; of these 10 (7%) had both a bone scan and a PET-CT scan. Bone scans and PET-CT scans were interpreted by nuclear medicine subspecialty trained radiologist. In addition to plain radiographs of the involved bone, CT and/or MRI of the involved bone, as well as chest CT were routinely obtained in all patients as part of their work-up.
The cohort included 87 males and 51 females with mean age 54±20 years. Tumor location included pelvis/sacrum (n=49, 36%), femur (n=33, 24%), humerus (n=14, 10%), spine (n=12, 9%), scapula (n=9, 7%), sternum (n=9, 7%), ribs (n=8, 6%), radius (n=1, 1%), tibia (n=1, 1%), fibula (n=1, 1%), and calcaneus (n=1, 1%). Eighteen (13%) patients presented with pathologic fracture. Mean tumor size and volume were 10±6 cm and 513±939 cm3, respectively. Tumor grade included Grade 1 (n=50, 36%), Grade 2 (n=37, 27%), Grade 3 (n=17, 12%), dedifferentiated (n=26, 19%), clear cell (n=6, 4%), and mesenchymal (n=2, 1%). All tumors were graded by subspecialty trained bone and soft-tissue pathologists based on the Lichtenstein and Jaffe scoring system (10). The institution approved the human protocol for this investigation and all investigations were conducted in conformity with the ethical principles of research.
Statistical analysis. Unpaired Student t-tests were used to assess continuous variables and the Fisher exact test was used to compare categorical variables. Values are reported as mean±standard deviation (SD). The sensitivity and specificity, as well as the positive and negative predictive value of bone scintigraphy and PET-CT were calculated and reported. A p-value of <0.05 was considered significant.
Results
Histology comparison. At the time of presentation, patients with dedifferentiated chondrosarcoma (66±12 years) were older (p<0.001) compared to patients with Grade 1 (47±23 years), Grade 2 (57±16 years), Grade 3 (60±18), clear cell (34±15 years) and mesenchymal (29±8 years) chondrosarcoma. There was no difference (p=0.36) in the proportion of males in each group based on histology; Grade 1 (n=29, 58%), Grade 2 (n=28, 76%), Grade 3 (n=9, 53%), dedifferentiated (n=16, 62%), clear cell (n=3, 50%) and mesenchymal (n=2, 100%) chondrosarcoma.
Patients with dedifferentiated chondrosarcoma (13±7 cm) were noted to have larger tumors (p<0.001) in terms of maximal tumor dimension compared to Grade 1 (7±5 cm), Grade 2 (10±6 cm), Grade 3 (11±4 cm), clear cell (4±1 cm) and mesenchymal (6±1 cm) chondrosarcoma. However, there was no difference (p=0.38) in the mean tumor volume comparing Grade 1 (359±802 cm3), Grade 2 (700±1,362 cm3), Grade 3 (679±770 cm3), dedifferentiated (578±577 cm3), clear cell (34±21 cm3) and mesenchymal (154±108 cm3) chondrosarcoma. Patients with dedifferentiated chondrosarcoma (n=9, 35%) were more likely to present with a pathological fracture (p=0.01) compared to Grade 1 (n=4, 8%), Grade 2 (n=2, 5%), Grade 3 (n=3, 18%), clear cell (n=0, 0%) and mesenchymal (n=0, 0%) chondrosarcoma.
Bone metastatic disease at presentation. Overall, 17 (12%) patients had a positive bone scintigraphy or PET-CT for distant skeletal metastases (Table I). Of these, 8 were confirmed to be metastatic disease and 9 were interpreted as false positives after further investigation. The overall sensitivity and specificity of a bone scan or PET-CT to evaluate metastatic bone disease in patients with chondrosarcoma was 100% (95%CI=63.1-100%) and 93.1% (95%CI=87.3-96.8%), respectively. The positive and negative predictive values were 47.1% (95%CI=32.1-62.5%) and 100%; with an accuracy of 93.5% (95%CI=87.9-96.9%).
In the patients with a positive bone scintigraphy (n=11), 6 proved to be benign after further investigation, with etiologies including osteochondroma (n=1), benign bone island (n=1), fibrous dysplasia (n=1), stress reaction (n=1), post-traumatic deformity (n=1), and no identifiable bone lesion on MRI (n=1). The remaining 5 patients had biopsy proven skeletal metastases, all in the setting of dedifferentiated chondrosarcoma. For a bone scan, the sensitivity was 100% (95%CI=47.8-100%) and the specificity was 94.7% (95%CI=88.9-98.0%); with a positive and negative predictive value of 45.5% (95%CI=27.6-64.5%) and 100%, respectively. The accuracy of a bone scan was 94.9% (95%CI=89.4-98.1%).
In the patients with a PET-CT, 6 were read as having skeletal metastases. In 3 cases biopsies were consistent with metastatic disease; with all cases occurring in the setting of a dedifferentiated chondrosarcoma. In the remaining 3 cases, one was felt to be a benign enchondroma on serial imaging, one patient was found to have metastatic prostate cancer, and in one patient the biopsy was negative. The sensitivity was 100% (95%CI= 29.2-100%) and the specificity was 88.5% (95%CI=69.9-97.6%); with a positive and negative predictive value of 50% (95CI=25.6-74.4%) and 100%, respectively. The accuracy of a PET-CT was 89.7% (95%CI=72.7-97.8%).
In the patients with a bone scan and a PET-CT, there were no cases of discrepancy in the results between scans. In addition, no patients were found to develop osseous metastasis within the first 2-years of follow-up. Likewise, in the 138 patients who did not undergo a bone scintigraphy or PET/CT, none were found to have isolated bone metastatic disease at a mean follow-up of 10±5 years.
Patients with dedifferentiated chondrosarcoma (n=8, 31%) were more likely to have bone metastatic disease (p<0.001) at presentation compared to Grade 1 (0%), Grade 2 (0%), Grade 3 (0%), clear cell (0%) and mesenchymal (0%) chondrosarcoma. Patients with bone metastatic disease were more likely to have larger tumors in terms of size (14±4 vs. 9±5 cm, p=0.03) and volume (888±1,048 vs. 490±931 cm3, p=0.24). In addition, patients presenting with a pathologic fracture were more likely to present with bone metastatic disease compared to those without a pathological fracture (n=6, 33% vs. n=2, 2%; p<0.001).
Discussion
Chondrosarcoma is a common primary malignant bone tumor. Traditionally, surgical staging for chondrosarcomas has included bone scintigraphy or PET-CT for exclusion of skeletal metastases. The results of the current study question the utility of this practice, and favor a selective use of bone scans only in the setting of dedifferentiated histology.
Bone metastases can occur in chondrosarcoma but are typically seen in the setting of recurrent disease and/or in conjunction with pulmonary metastases (11-13). Isolated skeletal metastases at presentation of a primary tumor are exceedingly rare and have frequently been described in the setting of high grade or dedifferentiated disease (5-7, 14, 15). In a recent large series, Gulia et al. (5) only noted bone metastases in patients with Grade 2 chondrosarcoma, and as such only recommend staging studies for chondrosarcoma to include a non-contrast CT scan of the chest. Unlike the study by Gulia et al. (5), we identified eight patients in our study group with isolated skeletal metastases at presentation, all occurred in the setting of dedifferentiated histology. High grade and dedifferentiated lesions are known to be very aggressive tumors with associated high rates of local recurrence, metastases, and death, with a 5 year over-all survival rate of 10.5% (2, 16). It is worth noting that of all patients who did not have staging with bone scintigraphy or PET/CT, none went on to develop skeletal metastasis which can likely be attributed to the relative rarity of this condition in conventional chondrosarcoma of bone.
Bone scintigraphy has relatively low positive predictive value for detection of skeletal metastases with multiple benign osseous lesions as well as degenerative disease often read as metastases leading to unnecessary further testing and delays in treatment (8). Eleven patients in our group had staging scintigraphy read as distant metastases. Of these, six proved to be false positives on further investigation. Increased uptake on bone scintigraphy as part of staging for skeletal chondrosarcoma, has been noted to be common, with one series reporting uptake in 1/3 of patients, with none of the areas being positive for metastatic disease (6). The utility of bone scintigraphy as part of routine surgical staging in patients with chondrosarcoma of the bone is often not necessary for histologies other than dedifferentiated chondrosarcoma. As such we would only recommend the ordering of this imaging by orthopedic oncologists and not by referring orthopedic surgeons or primary care physicians to evaluate a cartilage tumor.
Recent reports have demonstrated PET-CT to have higher sensitivity and specificity for detecting skeletal metastases in primary bone sarcomas (17-19). Furthermore, FGD-PET/CT has been shown to be a useful adjunct in the evaluation of chondrosarcoma patients, not only in the identification of local recurrence and/or metastases, but also as a parameter for determination of benign vs malignant lesions, tumor grade, and prediction of outcomes (20-22). Nineteen patients in our cohort underwent PET-CT rather than bone scintigraphy for surgical staging, with 3 patients having bone metastases, leading to a slightly higher positive predictive value of PET-CT compared to bone scintigraphy for detecting bone metastatic disease, however the accuracy of the test was only 89.7%.
We acknowledge limitations to the current series. The retrospective nature of the study limits collection of data and the analysis performed. However, prospective collection of data by our institutional sarcoma registry helps minimize recall and selection bias. Additionally, the data were collected from a single institution with multiple treating orthopedic oncologists with initial surgical staging performed at their discretion, likely guided by various patient and tumor factors that are not accounted for in this study. The sample size is small and it is possible with larger numbers, the results of the study could be different.
Skeletal metastases at presentation in chondrosarcoma of the bone are rare. Furthermore, bone scintigraphy and PET/CT lack positive predictive value and may lead to unnecessary testing and delays in treatment. It is our recommendation that bone scintigraphy and PET/CT are not routinely needed in the surgical staging of patients with presumed low or intermediate grade lesions. Rather, we advocate for an individualized approach specific to each patient case with consideration for bone scintigraphy or PET/CT in the setting of dedifferentiated and other high grade lesions as these patients are at increased risk for presentation with metastases.
Acknowledgements
No disclosures of funding were received for this work from NIH, Wellcome Trust, or HHMI.
Footnotes
Authors' Contributions
Johnson: Drafting of initial and final manuscript, data collection, data analysis; Rainer: Drafting of initial and final manuscript, data collection, data analysis; Rose: Review and editing of final manuscript; Houdek: Drafting of initial and final manuscript, data analysis, supervision.
Conflicts of Interest
There are no conflicts of interest related to this study.
- Received August 4, 2020.
- Revision received August 30, 2020.
- Accepted August 31, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved