**9.1 Benign bone tumors and tumor- like disorders**

In benign bone tumors, the uptake of radiopharmaceuticals (e.g. 99mTc) varies by type of tumor and may be normal, mild or severe. Bone scintigraphy is useful for the diagnosis of osteoid osteoma, especially when it is located at the spine, pelvis or hip, where radiological studies are usually not diagnosed. The typical scintigraphic finding is a round focal uptake lesion. CT is always necessary to confirm the diagnosis and surgical treatment. Most enchondromas appear like hot spots at bone scan study. This technique can locate these tumors in multiple enchondromatosis, but it can not differentiate between enchondroma and chondrosarcoma. The scintigraphic feature of the giant cell tumor is increased tracer uptake in all phases of this study, and the image of donut of the lesion, with a rim of uptake surrounding a central area of low uptake. Nevertheless, this image may also appear in other bone tumors. In fibrous displasia, characterized by replacement of normal bone tissue by abnormal fibro-osseous tissue with a high bone turnover, bone scan also display areas of increased uptake. In other bone lesions such as Langerhans cell histiocytosis, hemangiomas and aneurysmal bone cysts, the sensitivity of this technique is variable (Schneider, 2006).

### **9.2 Soft tissue tumors and primary malignant bone tumors**

Musculoskeletal sarcomas represent a heterogeneous group of malignancies involving bone and soft tissue. Multiple myeloma is the most common primary malignancy of bone in adults, with an incidence of 3 per 100,000 in the USA. It may affect any bone with hematopoietic red marrow. Patients affected are usually over 50 years of age with the most common age group being between 60 and 65 years of age. Excluding myeloma and lymphoma, malignant primary bone tumors constitute only 0.2% of all malignancies in adults and approximately 5% of childhood malignancies, and, excluding mieloma in adults, the overwhelming majority of cases consist of osteosarcoma or Ewings' sarcoma (Green, 2009). Both are more common in the pediatric than the adult population. Osteosarcoma is the most frequent primary bone malignancy in children and second in adults following multiple myeloma. The Ewing's sarcoma family of tumors is the second most frequent primary bone malignancy in children and young adults and it is the most lethal bone tumor. The most common presenting symptom for primary bone tumors is a painful swelling arising in the bone. The presentation may be similar to acute or chronic osteomyelitis with systemic symptoms of fever, malaise, weight loss, and leukocytosis. Approximatly 15% of patients have clinically evident metastasic disease at diagnosis. Metastatic spread is mainly hematogenous, and the lungs are the most common site of metastases, followed by bone and bone marrow. Anatomic imaging techniques including radiography, US, CT and MRI, currently play a dominant role in the evaluation of suspected and known sarcomas of both soft tissue and bone. Nuclear medicine techniques such as scintigraphy, Thallium-201, and 67Ga imaging have all been used in the assessment of primary bone tumors. However, PET

Nuclear Medicine in Musculoskeletal Disorders: Clinical Approach 115

the information from three-phase scintigraphy (88%), compared with static imaging alone (74%) where the blood flow and blood pool images showed a reduction in vascularity and extension. These are consisting features with the results published in previous studies (Cook et al., 2010; as cited in Knop et al., 1990, and Sommer et al., 1987). The sensitivity of 18F-FDG-PET in staging primary bone tumors appears to vary between different tumor types and location of metastases. Spiral CT is the modality of choice for detection of relatively small lung metastases. Franzius et al compared the detection lung metastases of sarcoma by CT and 18F-FDG-PET, showing a higher sensitivity for the former modality especially in lesions <9 mm. 18F-FDG-PET may, however, assist in differentiating nonspecific lung nodules from metastases detected by CT in case of larger lung lesions, within the size range of PET resolution. 18F-FDG-PET may also identify unexpected extra pulmonary metastases. Serial 18F-FDG-PET assessment of primary bone tumors (predominantly osteosarcomas, Ewing's sarcomas, or both) is a good non-invasive method to predict pathologic neoadjuvant chemotherapy response (Cook et al., 2010; Even-Sapir, 2007). Another earlier study also showed a correlation with pathologic response but described high 18F-FDG uptake in granulation and/or fibrotic tissue and in the fibrous pseudocapsule of treated tumors (Cook

Multiple myeloma hardly triggers osteoblastic reaction and therefore scintigraphy is less sensitive than plain radiography and CT. FDG-PET indicates active myeloma and CT shows bone destruction. Therefore hybrid whole-body PET/CT is an excellent method to evaluate myeloma. Currently whole-body and spinal MRI and PET/CT are considered the imaging techniques of choice for initial evaluation and follow-up of these patients. Durie et al assessed the role of 18F-FDG-PET in 66 patients with multiple myeloma and monoclonal gammopathy of undetermined significance. Their results suggested that a positive 18F-FDG-PET reliably indicates the presence of active myeloma, whereas a negative study strongly supports the diagnosis of monoclonal gammopathy of undetermined significance. 18F-FDG-PET has also been reported to identify unexpected medullar and extramedullar sites of myelomatous disease not appreciated on X-ray, CT, or scintigraphy (Even-Sapir, 2007). In a recent report on 28 patients with multiple myeloma, 18F-FDG-PET/CT and MRI of the spine were shown to have complementary roles. Although the former modality detected more lesions, all of which were located outside the field of view of MRI, the latter modality was found superior for diagnosing an infiltrative pattern in the spine (Nanni et al., 2006). In another recent report, 18F-FDG-PET was found to be valuable in detecting infection in

Osteosarcoma represents only 0.1% of all tumors, but it is the second most frequent malignant primary bone tumor after myeloma. The diagnosis of osteosarcoma is based on characteristic histologic features in combination with typical radiographic findings. MRI of the entire suspected bone is performed to define the degree of penetration of the tumor surrounding soft tissue as well as to estimate the local tumor infiltration into bone marrow. Furthermore, CT of the chest and conventional bone scan are necessary for early detection of metastases. MRI and scintigraphy are also used to distinguish postoperative changes from residual or recurrent tumor tissue after local surgical treatment. Because osteosarcoma metastases usually incorporate bisphosphonates, bone scanning can be used for follow-up examinations to detect both osseous and nonosseous metastases. High-resolution CT has been shown to be superior to 18F-FDG-PET for detecting lung metastases. Data on the benefit of 18F-FDG-PET for detecting skeletal metastases in osteosarcoma patients are still very sparse, but successful detection of all sites of bone involvement by 18F-FDG-PET has

et al., 2010; as cited in Jones et al., 1996).

patients with multiple myeloma (Even-Sapir, 2007).

is becoming the most imporant modality for assessing biologic characteristics of the tumor, for primary staging, and for determining response to treatment. Although imaging studies may be highly suggestive of the diagnosis, they cannot reliably differentiate among the various types of malignant bone tumors, and even among malignant and benign conditions. Histopathologic confirmation, therefore, is required. The sites for biopsy are critical for accurate histological diagnosis and staging because biopsy of a small site may not represent the overall character of the tumor, missing high-grade areas, and non-diagnostic biopsies may also occur (Howman-Giles et al., 2006).

The diagnosis of indeterminate bone lesions is limited with 18F-FDG-PET, but in general the greater the level of uptake, the more likely a lesion is malignant in nature. However, it has been reported that some giant cell tumors and fibrous dysplasia may show uptake equivalent to osteosarcomas and that some other benign bone lesions may show high 18F-FDG accumulation (Aoki et al., 2001). Despite this, 18F-FDG-PET has a high specificity for excluding malignant bone tumors (Cook et al., 2010). Recently, dual time point imaging and calculation of a retention index for 18F-FDG have shown improved discrimination of benign and malignant bone lesions compared with static measures, but that some overlap was still present (Tian et al., 2009). Lodge et al observed difference in time-activity curves between benign and low-grade malignant tumors that show peak activity within the first 30 minutes post-injection and high-grade sarcomas, which reach peak activity 4 hours after injection. This quantitative approach cannot separate low-grade sarcomas from benign lesions. MRI is the modality of choice to define the extension of tumors to surrounding soft tissue as well as to estimate the local tumor infiltration into bone marrow. However, in the pediatric population, 18F-FDG-PET is valuable for detection of skip metastases in cases of equivocal MRI findings due to the physiological red blood marrow in long bones (Even-Sapir, 2007; as cited in Wuisman P, Enneking WF, 1990). With the new hybrid imaging, it is now possible to take advantage of the metabolic and morphologic information from 18F-FDG-PET/CT to enhance discrimination between benign and malignant bone lesions by dedicated interpretation of the CT characteristics (Strobel et al., 2008). 18F-FDG-PET data can also assist in optimizing the biopsy site of heterogeneous masses by guiding sampling to active tumor sites and avoiding errors due to biopsy of necrotic tumor areas (Even-Sapir, 2007; as cited in Pezeshk et al., 2006). In general, 18F-FDG-PET or PET/CT would appear to have a complementary role to conventional staging procedures (Cook et al., 2010; as cited in Kleis et al., 2009, Kneisl et al., 2006 and Völker et al., 2007). After a diagnosis of a malignant primary bone lesion is made, the use of bone scintigraphy to define the extent of tumor before surgical resection is controversial: good correlation between increased bone tracer uptake and true anatomical extent that has been reported (Cook et al., 2010; as cited in Goldmann et al., 1975, McKillop et al., 1981, and Papanicolou et al., 1982), has not been supported by other studies (Chew & Hudson, 1982). These discrepancies may be due to peritumoral reactive changes overestimating extent or underestimations due to inability to detect marrow and soft-tissue involvement. For these reasons, MRI is the most accurate noninvasive assessment of tumor extent. On the other hand, several studies have reported the ability of bone scintigraphy to predict histological response to preoperative chemotherapy in patients with primary malignant bone tumors. Ozcan et al have reported a study of 27 patients with osteosarcoma, Ewing's sarcoma, and malignant fibrous histiocytoma, which has displayed a reduction in hyperemia and extension as the most notable findings on three-phase bone scintigraphy. A reduction in tumor blood flow of 58.7% was found in 15 responding patients compared with 19.9% in the nonresponders. A higher accuracy in assessing response was possible using all

is becoming the most imporant modality for assessing biologic characteristics of the tumor, for primary staging, and for determining response to treatment. Although imaging studies may be highly suggestive of the diagnosis, they cannot reliably differentiate among the various types of malignant bone tumors, and even among malignant and benign conditions. Histopathologic confirmation, therefore, is required. The sites for biopsy are critical for accurate histological diagnosis and staging because biopsy of a small site may not represent the overall character of the tumor, missing high-grade areas, and non-diagnostic biopsies

The diagnosis of indeterminate bone lesions is limited with 18F-FDG-PET, but in general the greater the level of uptake, the more likely a lesion is malignant in nature. However, it has been reported that some giant cell tumors and fibrous dysplasia may show uptake equivalent to osteosarcomas and that some other benign bone lesions may show high 18F-FDG accumulation (Aoki et al., 2001). Despite this, 18F-FDG-PET has a high specificity for excluding malignant bone tumors (Cook et al., 2010). Recently, dual time point imaging and calculation of a retention index for 18F-FDG have shown improved discrimination of benign and malignant bone lesions compared with static measures, but that some overlap was still present (Tian et al., 2009). Lodge et al observed difference in time-activity curves between benign and low-grade malignant tumors that show peak activity within the first 30 minutes post-injection and high-grade sarcomas, which reach peak activity 4 hours after injection. This quantitative approach cannot separate low-grade sarcomas from benign lesions. MRI is the modality of choice to define the extension of tumors to surrounding soft tissue as well as to estimate the local tumor infiltration into bone marrow. However, in the pediatric population, 18F-FDG-PET is valuable for detection of skip metastases in cases of equivocal MRI findings due to the physiological red blood marrow in long bones (Even-Sapir, 2007; as cited in Wuisman P, Enneking WF, 1990). With the new hybrid imaging, it is now possible to take advantage of the metabolic and morphologic information from 18F-FDG-PET/CT to enhance discrimination between benign and malignant bone lesions by dedicated interpretation of the CT characteristics (Strobel et al., 2008). 18F-FDG-PET data can also assist in optimizing the biopsy site of heterogeneous masses by guiding sampling to active tumor sites and avoiding errors due to biopsy of necrotic tumor areas (Even-Sapir, 2007; as cited in Pezeshk et al., 2006). In general, 18F-FDG-PET or PET/CT would appear to have a complementary role to conventional staging procedures (Cook et al., 2010; as cited in Kleis et al., 2009, Kneisl et al., 2006 and Völker et al., 2007). After a diagnosis of a malignant primary bone lesion is made, the use of bone scintigraphy to define the extent of tumor before surgical resection is controversial: good correlation between increased bone tracer uptake and true anatomical extent that has been reported (Cook et al., 2010; as cited in Goldmann et al., 1975, McKillop et al., 1981, and Papanicolou et al., 1982), has not been supported by other studies (Chew & Hudson, 1982). These discrepancies may be due to peritumoral reactive changes overestimating extent or underestimations due to inability to detect marrow and soft-tissue involvement. For these reasons, MRI is the most accurate noninvasive assessment of tumor extent. On the other hand, several studies have reported the ability of bone scintigraphy to predict histological response to preoperative chemotherapy in patients with primary malignant bone tumors. Ozcan et al have reported a study of 27 patients with osteosarcoma, Ewing's sarcoma, and malignant fibrous histiocytoma, which has displayed a reduction in hyperemia and extension as the most notable findings on three-phase bone scintigraphy. A reduction in tumor blood flow of 58.7% was found in 15 responding patients compared with 19.9% in the nonresponders. A higher accuracy in assessing response was possible using all

may also occur (Howman-Giles et al., 2006).

the information from three-phase scintigraphy (88%), compared with static imaging alone (74%) where the blood flow and blood pool images showed a reduction in vascularity and extension. These are consisting features with the results published in previous studies (Cook et al., 2010; as cited in Knop et al., 1990, and Sommer et al., 1987). The sensitivity of 18F-FDG-PET in staging primary bone tumors appears to vary between different tumor types and location of metastases. Spiral CT is the modality of choice for detection of relatively small lung metastases. Franzius et al compared the detection lung metastases of sarcoma by CT and 18F-FDG-PET, showing a higher sensitivity for the former modality especially in lesions <9 mm. 18F-FDG-PET may, however, assist in differentiating nonspecific lung nodules from metastases detected by CT in case of larger lung lesions, within the size range of PET resolution. 18F-FDG-PET may also identify unexpected extra pulmonary metastases. Serial 18F-FDG-PET assessment of primary bone tumors (predominantly osteosarcomas, Ewing's sarcomas, or both) is a good non-invasive method to predict pathologic neoadjuvant chemotherapy response (Cook et al., 2010; Even-Sapir, 2007). Another earlier study also showed a correlation with pathologic response but described high 18F-FDG uptake in granulation and/or fibrotic tissue and in the fibrous pseudocapsule of treated tumors (Cook et al., 2010; as cited in Jones et al., 1996).

Multiple myeloma hardly triggers osteoblastic reaction and therefore scintigraphy is less sensitive than plain radiography and CT. FDG-PET indicates active myeloma and CT shows bone destruction. Therefore hybrid whole-body PET/CT is an excellent method to evaluate myeloma. Currently whole-body and spinal MRI and PET/CT are considered the imaging techniques of choice for initial evaluation and follow-up of these patients. Durie et al assessed the role of 18F-FDG-PET in 66 patients with multiple myeloma and monoclonal gammopathy of undetermined significance. Their results suggested that a positive 18F-FDG-PET reliably indicates the presence of active myeloma, whereas a negative study strongly supports the diagnosis of monoclonal gammopathy of undetermined significance. 18F-FDG-PET has also been reported to identify unexpected medullar and extramedullar sites of myelomatous disease not appreciated on X-ray, CT, or scintigraphy (Even-Sapir, 2007). In a recent report on 28 patients with multiple myeloma, 18F-FDG-PET/CT and MRI of the spine were shown to have complementary roles. Although the former modality detected more lesions, all of which were located outside the field of view of MRI, the latter modality was found superior for diagnosing an infiltrative pattern in the spine (Nanni et al., 2006). In another recent report, 18F-FDG-PET was found to be valuable in detecting infection in patients with multiple myeloma (Even-Sapir, 2007).

Osteosarcoma represents only 0.1% of all tumors, but it is the second most frequent malignant primary bone tumor after myeloma. The diagnosis of osteosarcoma is based on characteristic histologic features in combination with typical radiographic findings. MRI of the entire suspected bone is performed to define the degree of penetration of the tumor surrounding soft tissue as well as to estimate the local tumor infiltration into bone marrow. Furthermore, CT of the chest and conventional bone scan are necessary for early detection of metastases. MRI and scintigraphy are also used to distinguish postoperative changes from residual or recurrent tumor tissue after local surgical treatment. Because osteosarcoma metastases usually incorporate bisphosphonates, bone scanning can be used for follow-up examinations to detect both osseous and nonosseous metastases. High-resolution CT has been shown to be superior to 18F-FDG-PET for detecting lung metastases. Data on the benefit of 18F-FDG-PET for detecting skeletal metastases in osteosarcoma patients are still very sparse, but successful detection of all sites of bone involvement by 18F-FDG-PET has

Nuclear Medicine in Musculoskeletal Disorders: Clinical Approach 117

between the osteoclastic and osteoblastic processes, the radiographic appearance of a bone metastasis may be lytic, sclerotic (blastic), or mixed. The osteoblastic component of the metastasis represents reaction of normal bone to the metastatic process. The incidence of lytic, blastic, and mixed types of bone metastases is different in various tumor types. Lytic lesions may be seen in almost all tumor types. Bone metastases of bladder, kidney, and thyroid cancer and lesions of multiple myeloma are invariably lytic. Blastic lesions are frequently seen in prostate and breast cancer, occasionally in lung, stomach, pancreas, and cervix carcinomas, and infrequently in colorectal cancer (Beheshti et al., 2009; Even-Sapir, 2005). The most frequent distribution of metastasis in the human skeleton is usually 80% in the axial skeleton and ribs, 10% skull and 10% in long bones. Approximately 40% of patients with metastases have no pain at diagnosis (Diaz & De Haro, 2005). Symptoms occur mainly when the lesion increases in size, causing extensive bone destruction, which may lead to collapse or fracture, or in the presence of accompanying complications, such as spinal cord

Bone scan is the primary tool for screening or monitoring bone metastases due to its high sensitivity, versus plain radiography (Brown, 1993), and plays an integral part in tumor staging and management, since early detection of skeletal metastases optimizes management (Even-Sapir, 2005). Scintigraphic image of metastases is one or more high uptake foci in 98% of the cases, and the usual pattern consists of increased radiotracer deposition in areas of osteoblastic reparative activity in response to tumor osteolysis. The presence of multiple, randomly, distributed areas of increased uptake of varying size, shape, and intensity is highly suggestive of bone metastases, specially at sternum, scapula and ribs. Although multiple foci of increased activity may be encountered in other pathologic conditions, it is often possible to distinguish metastatic disease from other entities by analyzing the pattern of distribution of the abnormalities. Traumatic injury, in contrast to metastatic disease, generally manifests as discrete focal abnormalities of similar intensity. Multifocal rib trauma has a characteristic linear distribution. In patients with osteoporosis, the presence of kyphosis and/or an H-shaped sacral fracture suggests the correct diagnosis. In older patients, osteoarthritis and degenerative changes may manifest as areas of intense activity on radionuclide bone images. These changes can be distinguished from metastatic disease by virtue of their characteristic location (eg, knees, hands, wrists). Involvement of both sides of the joint is common in arthritis but unusual in malignant conditions. The remaining 2% is mild uptake foci owing to preponderance of osteolytic activity. The possibility of an artifact should be ruled out in cases of well-defined cold spot. In cases of wide-spread metastases, the radiopharmaceutical can be almost completely captured and it may lead to a superscan image, where kidney or bladder silhouettes are not initially seen by the delay in urinary excretion of radiotracer. A superscan may also be associated with metabolic bone disease but, in this case, the uptake is more uniform in appearance and extends into the distal appendicular skeleton. Intense calvarial uptake that is disproportionate to that in the remainder of the skeleton is another feature of a metabolic superscan. SPECT is reported to detect 20 to 50% more lesions in the spine compared with planar scintigraphy, and it increases both the sensitivity and specificity. The new hybrid system, SPECT with multislice CT, improves diagnostic accuracy (Dasgeb et al., 2007). The most common radiotracer is Tc-99m-MDP but in patients with follicular thyroid carcinoma or in cases of neuroblastoma, iodine-131 (131I) and 123I-Metaiodobenzylguanidine (123I-MIBG) are more sensitive. Increased uptake of these radiotracers reflects the osteoblastic reaction of bone to the destruction of bone by the tumor cells, whereas increased 18F-FDG activity at the sites of bone

compression or nerve root invasion.

been reported recently (Even-Sapir, 2007; as cited in Franzius et al., 2002). Nevertheless, in children there may be an exception for primary staging, where there may be an indication for 18F-FDG-PET to detect intraosseous skip metastases in cases of unequivocal MRI findings, although no data are yet available to support this hypothesis. PET scans will not obviate the need for biopsy and tissue diagnosis in soft-tissue and bone masses, but it is remarkably helpful to guide biopsy. Non-PET–guided biopsy might miss the most biologically significant region, resulting in a false low pre-therapeutic tumor grading. 18F-FDG-PET imaging data has shown reliability for prediction of tumor response to preoperative, neoadjuvant chemotherapy. On the other hand, for differentiation between benign residual mass lesions caused by post-therapeutic tissue changes and residual tumor tissue or local relapse, 18F-FDG PET is considered to be highly sensitive and more accurate than CT or MRI (Brenner et al., 2003). A high baseline uptake of 18F-FDG in osteosarcoma has been reported as showing an inverse correlation with prognostic indicators and is associated with a poor outcome with similar results for patients with high post-treatment FDG activity (Cook et al., 2010; as cited in Costelloe et al., 2009; Franzius et al., 2002).

Ewing's sarcoma is a highly malignant primary bone tumor that is being derived from red bone marrow. It accounts for approximately 5% of biopsy-analyzed bone tumors and approximately 33% of primary bone tumors. No single morphologic or functional imaging method provides findings for a specific diagnosis of Ewing's sarcoma, but the results do contribute to tumor staging. Because the clinical symptoms of Ewing's sarcoma are nonspecific and because they frequently suggest osteomyelitis, an initial conventional radiographic and/or MRI examination is performed. With static bone scintigraphy, Ewing's sarcoma is usually depicted as a focal area of increased radionuclide activity. Whole-body bone scans can provide information about the primary lesion and depict skip lesions. Also, bone scintigraphy can be used to localize distant metastases during tumor staging. Three-phase dynamic bone scintigraphy can help in the assessment of treatment effects, with a reported accuracy of 88%. In cases that respond to treatment, a reduction of both flow and tracer uptake can be observed. 18F-FDG-PET may help to detect lesions that are not shown on conventional bone scans. It is the most sensitive modality for therapeutic follow-up, and this modality can reveal early changes in tumor metabolism, which is an indicator of the therapeutic effect.

Regarding the lymphomatous disease, primary skeletal involvement occurs in 3 to 5% of patients with non-Hodgkin's lymphoma, and secondary bone involvement occurs in up to 25% of patients. Moog et al have reported 18F-FDG-PET to be more sensitive and specific than Tc-99m-MDP bone scintigraphy for detection of osseous involvement by lymphoma. Early bone involvement may present as abnormal on 18F-FDG-PET with normal CT appearance, since detection of malignant bone involvement on CT depends on the presence of a considerable amount of bone destruction.

#### **9.3 Metastases**

Bone metastasis is the most common malignant bone tumor. It affects two thirds of cancer patients, and tumors that most often lead to metastases are breast, lung and prostate neoplasms. Bone involvement by cancer occurs most commonly by hematogenous spread, although tumor may occasionally extend directly from the soft tissue to the adjacent bone. The vast majority of bone metastases initiate as intramedullary lesions. The normal bone undergoes constant remodeling, maintaining a balance between osteoclastic (resorptive) and osteoblastic activity. As the metastatic lesion enlarges within the marrow, the surrounding bone undergoes osteoclastic and osteoblastic reactive changes. Based on the balance

been reported recently (Even-Sapir, 2007; as cited in Franzius et al., 2002). Nevertheless, in children there may be an exception for primary staging, where there may be an indication for 18F-FDG-PET to detect intraosseous skip metastases in cases of unequivocal MRI findings, although no data are yet available to support this hypothesis. PET scans will not obviate the need for biopsy and tissue diagnosis in soft-tissue and bone masses, but it is remarkably helpful to guide biopsy. Non-PET–guided biopsy might miss the most biologically significant region, resulting in a false low pre-therapeutic tumor grading. 18F-FDG-PET imaging data has shown reliability for prediction of tumor response to preoperative, neoadjuvant chemotherapy. On the other hand, for differentiation between benign residual mass lesions caused by post-therapeutic tissue changes and residual tumor tissue or local relapse, 18F-FDG PET is considered to be highly sensitive and more accurate than CT or MRI (Brenner et al., 2003). A high baseline uptake of 18F-FDG in osteosarcoma has been reported as showing an inverse correlation with prognostic indicators and is associated with a poor outcome with similar results for patients with high post-treatment

FDG activity (Cook et al., 2010; as cited in Costelloe et al., 2009; Franzius et al., 2002).

changes in tumor metabolism, which is an indicator of the therapeutic effect.

of a considerable amount of bone destruction.

**9.3 Metastases** 

Ewing's sarcoma is a highly malignant primary bone tumor that is being derived from red bone marrow. It accounts for approximately 5% of biopsy-analyzed bone tumors and approximately 33% of primary bone tumors. No single morphologic or functional imaging method provides findings for a specific diagnosis of Ewing's sarcoma, but the results do contribute to tumor staging. Because the clinical symptoms of Ewing's sarcoma are nonspecific and because they frequently suggest osteomyelitis, an initial conventional radiographic and/or MRI examination is performed. With static bone scintigraphy, Ewing's sarcoma is usually depicted as a focal area of increased radionuclide activity. Whole-body bone scans can provide information about the primary lesion and depict skip lesions. Also, bone scintigraphy can be used to localize distant metastases during tumor staging. Three-phase dynamic bone scintigraphy can help in the assessment of treatment effects, with a reported accuracy of 88%. In cases that respond to treatment, a reduction of both flow and tracer uptake can be observed. 18F-FDG-PET may help to detect lesions that are not shown on conventional bone scans. It is the most sensitive modality for therapeutic follow-up, and this modality can reveal early

Regarding the lymphomatous disease, primary skeletal involvement occurs in 3 to 5% of patients with non-Hodgkin's lymphoma, and secondary bone involvement occurs in up to 25% of patients. Moog et al have reported 18F-FDG-PET to be more sensitive and specific than Tc-99m-MDP bone scintigraphy for detection of osseous involvement by lymphoma. Early bone involvement may present as abnormal on 18F-FDG-PET with normal CT appearance, since detection of malignant bone involvement on CT depends on the presence

Bone metastasis is the most common malignant bone tumor. It affects two thirds of cancer patients, and tumors that most often lead to metastases are breast, lung and prostate neoplasms. Bone involvement by cancer occurs most commonly by hematogenous spread, although tumor may occasionally extend directly from the soft tissue to the adjacent bone. The vast majority of bone metastases initiate as intramedullary lesions. The normal bone undergoes constant remodeling, maintaining a balance between osteoclastic (resorptive) and osteoblastic activity. As the metastatic lesion enlarges within the marrow, the surrounding bone undergoes osteoclastic and osteoblastic reactive changes. Based on the balance between the osteoclastic and osteoblastic processes, the radiographic appearance of a bone metastasis may be lytic, sclerotic (blastic), or mixed. The osteoblastic component of the metastasis represents reaction of normal bone to the metastatic process. The incidence of lytic, blastic, and mixed types of bone metastases is different in various tumor types. Lytic lesions may be seen in almost all tumor types. Bone metastases of bladder, kidney, and thyroid cancer and lesions of multiple myeloma are invariably lytic. Blastic lesions are frequently seen in prostate and breast cancer, occasionally in lung, stomach, pancreas, and cervix carcinomas, and infrequently in colorectal cancer (Beheshti et al., 2009; Even-Sapir, 2005). The most frequent distribution of metastasis in the human skeleton is usually 80% in the axial skeleton and ribs, 10% skull and 10% in long bones. Approximately 40% of patients with metastases have no pain at diagnosis (Diaz & De Haro, 2005). Symptoms occur mainly when the lesion increases in size, causing extensive bone destruction, which may lead to collapse or fracture, or in the presence of accompanying complications, such as spinal cord compression or nerve root invasion.

Bone scan is the primary tool for screening or monitoring bone metastases due to its high sensitivity, versus plain radiography (Brown, 1993), and plays an integral part in tumor staging and management, since early detection of skeletal metastases optimizes management (Even-Sapir, 2005). Scintigraphic image of metastases is one or more high uptake foci in 98% of the cases, and the usual pattern consists of increased radiotracer deposition in areas of osteoblastic reparative activity in response to tumor osteolysis. The presence of multiple, randomly, distributed areas of increased uptake of varying size, shape, and intensity is highly suggestive of bone metastases, specially at sternum, scapula and ribs. Although multiple foci of increased activity may be encountered in other pathologic conditions, it is often possible to distinguish metastatic disease from other entities by analyzing the pattern of distribution of the abnormalities. Traumatic injury, in contrast to metastatic disease, generally manifests as discrete focal abnormalities of similar intensity. Multifocal rib trauma has a characteristic linear distribution. In patients with osteoporosis, the presence of kyphosis and/or an H-shaped sacral fracture suggests the correct diagnosis. In older patients, osteoarthritis and degenerative changes may manifest as areas of intense activity on radionuclide bone images. These changes can be distinguished from metastatic disease by virtue of their characteristic location (eg, knees, hands, wrists). Involvement of both sides of the joint is common in arthritis but unusual in malignant conditions. The remaining 2% is mild uptake foci owing to preponderance of osteolytic activity. The possibility of an artifact should be ruled out in cases of well-defined cold spot. In cases of wide-spread metastases, the radiopharmaceutical can be almost completely captured and it may lead to a superscan image, where kidney or bladder silhouettes are not initially seen by the delay in urinary excretion of radiotracer. A superscan may also be associated with metabolic bone disease but, in this case, the uptake is more uniform in appearance and extends into the distal appendicular skeleton. Intense calvarial uptake that is disproportionate to that in the remainder of the skeleton is another feature of a metabolic superscan. SPECT is reported to detect 20 to 50% more lesions in the spine compared with planar scintigraphy, and it increases both the sensitivity and specificity. The new hybrid system, SPECT with multislice CT, improves diagnostic accuracy (Dasgeb et al., 2007). The most common radiotracer is Tc-99m-MDP but in patients with follicular thyroid carcinoma or in cases of neuroblastoma, iodine-131 (131I) and 123I-Metaiodobenzylguanidine (123I-MIBG) are more sensitive. Increased uptake of these radiotracers reflects the osteoblastic reaction of bone to the destruction of bone by the tumor cells, whereas increased 18F-FDG activity at the sites of bone

Nuclear Medicine in Musculoskeletal Disorders: Clinical Approach 119

Lesions that extend from the vertebral body into the posterior vertebral elements or involve the pedicle are more likely to represent metastases than lesions confined to the facet joints,

Osteoarthritis (OA) is the most prevalent chronic joint disease and it has the greatest health economic impact. Conventional radiography is still the first and most commonly used imaging technique for evaluation of a patient with a known or suspected diagnosis of OA (Guermazi, 2009). MRI is an appropriate tool for describing changes in cartilage volume and concomitant soft-tissue alterations. But for qualitative cartilage imaging, MRI has, to date, not been fully validated. Bone scan allows the differentiation of inflammatory from degenerative joint affections and may add information on the activity of the subchondral bone, which may develop to a prognostic marker of OA (Zacher et al., 2007). Another pronostic marker of slower progression that can help us deciding the most appropriate management is the imaging of the joints that show up as "cold" (Colamussi et al., 2004). Radionuclide joint imaging is more sensitive than clinical or radiographic techniques in detecting early joint involvement but usually it must be supplemented by other techniques

Usefulness of molecular imaging for early diagnosis of OA is still a challenge. Cartilage damage in OA is being recharacterized as having an earlier dynamic phase, where cartilage damage is potentially reversible, followed by an irreversible pathologic phase that ultimately leads to joint pain and immobility (Hu & Du, 2009). The point at which cartilage damage is deemed irreversible has not been defined but probably depends on the size of the lesion, age of the patient, underlying cause, comorbid factors, activity level, use of joint stabilizers, genetic predisposition, and other factors. To detect early cartilage damage, molecular imaging research has focused on the identification of better ways to either visualize extracellular matrix depletion or measure events that are associated with cartilage damage, such as chondrocyte death and the elaboration of matrix-degrading enzymes. In OA, there is general acceptance that abnormal chondrocyte apoptosis is a pivotal event in the eventual destruction of articular cartilage (Biswal et al., 2007). A method for the study of cell death in living subjets is based on an endogenous protein, annexin V, whose function is not clearly understood but which is thought to play a role in coagulation (Biswald et al., 2007; as cited in Reutelinsperger & Van Heerde, 1997). This protein has an extremely strong affinity for the cell membrane phospholipid phosphotidylserine, which is expressed to the outer surface of the cell membrane during the apoptotic cascade. The use of annexin V, labelled with either a radioisotope or a fluorescent marker, provides an excellent opportunity to image programmed cell death. To date, annexin V has been labelled with 99mTc, iodine (125I, 124I, 123I), 111In, 11C, gallium (Ga-67, Ga-68), and 18F, making it appropriate for either SPECT or PET imaging (Biswald et al., 2007; as cited in Blankenberg, 1998; Glaser, 2003; Lahorte, 2004; Russell, 2002; Zijlstra, 2003). However Annexin V imaging has yet to be applied to the assessment of human OA. Another event associated with cartilage damage is the elaboration of matrix-degrading enzymes. In OA damaged cartilage appears to activate hibernating proteases such as matrix metalloproteinases and cathepsins. Using a cathepsin B–sensitive near-infrared fluorescent probe, researchers have found significant amounts of signal arising from an arthritic knee compared with normal knees in

anterior vertebral body, or either side of a disc (Gnanasegaran et al., 2009).

to establish a specific diagnosis (Hoffer & Genant, 1976).

an animal model of OA (Biswald et al., 2007; as cited in Lai, 2004).

**10. Osteoarthritis** 

lesions on PET study represents active tumor itself. Bone scintigraphy is more sensitive than FDG-PET for detection of blastic/sclerotic lesion, whereas FDG-PET is more sensitive for lytic lesions and bone marrow disease. The latter has the additional ability to assess extraskeletal metastatic disease. Hybrid PET/CT imaging improves the specificity of FDG-PET for skeletal metastases (Dasgeb et al., 2007; as cited in Even-Sapir, 2005). An additional finding from scanning the peripheries, particularly in patients with bronchogenic carcinoma, may be the observation of hypertrophic osteoarthropathy secondary to cortical periostitis, that typically appears as symmetrical, nonuniform, irregular cortical uptake involving the long bones, most often seen in the femora, tibiae and wrists, and giving rise to the "tramline sign" (Gnanasegaran et al., 2009; Love et al., 2003; as cited in Ernstoff & Meehan, 2000). In patients with bone metastases who have received chemotherapy, reparative osteoblastic reaction that occurs after this treatment may lead to the appearance of bone areas with intense uptake during the first 3 months (flare phenomenon). As healing progresses, uptake in the lesion disminishes and by 6 months it should generally be possible to differentiate response from progression (Love et al., 2003).

18F-FDG-PET has become a routine imaging modality for staging and monitoring the response to therapy in patients with lymphoma. There are accumulating data indicating that 18F-FDG-PET may detect early marrow infiltration and may add clinically relevant information when performed in patients with primary or secondary lymphomatous bone involvement. FDG-PET can detect early marrow infiltration and therefore is more sensitive than planar scintigraphy or CT for assessment of early skeletal involvement in lymphoma (Dasgeb et al., 2007). A pattern of heterogeneous patchy marrow activity should raise the suspicion of marrow involvement in an 18F-FDG-PET study prior to therapy, while a pattern of diffuse uptake, mainly in Hodgkin's Lynphoma, is more commonly associated with reactive hematopoietic changes or myeloid hyperplasia (Even-Sapir, 2007).

Metastatic disease occasionally manifests as a solitary abnormality, usually in the spine, although other causes such as fractures, avascular osteonecrosis, primary bone tumors and infections must be previously ruled out. The location and/or characteristics of the lesion may guide the diagnostic suspicion but, especially if it is a solitary lesion, it must be studied with other imaging techniques such as CT or/and MRI (Schneider, 2006). Approximately 50% of cases in which scintigraphy detect a solitary focal uptake in a patient with a history of cancer, it is a metastasis. In patients with breast cancer, the sternum is a relatively common site to be affected often as a solitary lesion and probably results from local spread from the involved internal mammary lymph nodes. If a sternal lesion is situated distant from the manubriosternal junction, is irregular, asymmetric, or eccentric, then malignant involvement should be suspected. In a retrospective study of patients with breast cancer, 3.1% presented with an isolated sternal lesion and 76% of these were found to represent metastatic disease (Gnanasegaran et al., 2009; as cited in Kwai et al., 1988). Vertebral body fractures have a characteristic appearance on bone scintigraphy, showing a horizontal linear pattern of increased tracer accumulation. However, it is usually not possible to differentiate fractures due to benign diseases, such as osteoporosis from malignant collapse. In such cases, further evaluation with MRI is often the most informative. However, multiple linear abnormalities of varying intensity favour a benign etiology with presumed osteoporotic fracture occurring at different time points. Also, a follow-up bone scan after a few months that shows reducing activity at a vertebral fracture site, suggests a benign cause and a healing fracture. SPECT technique improves the localization and characterization of the vertebral lesions, due to its ability to delineate the body, pedicles, and spinous process: Lesions that extend from the vertebral body into the posterior vertebral elements or involve the pedicle are more likely to represent metastases than lesions confined to the facet joints, anterior vertebral body, or either side of a disc (Gnanasegaran et al., 2009).
